1695 (Author at Cato Institute) https://www.cato.org/ en Aaron Yelowitz discusses New York Medicaid budget woes on NPR’s All Things Considered https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitz-discusses-new-york-medicaid-budget-woes-nprs-all Fri, 24 Jan 2020 10:47:00 -0500 Aaron Yelowitz https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitz-discusses-new-york-medicaid-budget-woes-nprs-all New York City Is a Hot Spot for Illegal Medicaid Enrollment https://www.cato.org/publications/commentary/new-york-city-hot-spot-illegal-medicaid-enrollment Brian Blase, Aaron Yelowitz <div class="lead text-default"> <p>New York state is grappling with a&nbsp;Medicaid shortfall in the billions of dollars. And one of the main reasons is improper enrollment.</p> </div> , <div class="text-default"> <p>Using annual information from the Census Bureau to assess the demographic make‐​up of Medicaid enrollees over time, researcher Aaron Yelowitz and I&nbsp;estimated that 2.3 million to 3.3 million Medicaid enrollees nationally make an income in excess of what is allowed.</p> <p>This is of increasing importance given that ObamaCare massively expanded what was historically a&nbsp;welfare program for vulnerable populations like the disabled and low‐​income children and pregnant women — and tens of billions of taxpayer dollars are at stake.</p> <p>Excluding traditional pathways onto Medicaid (such as through disability or pregnancy), Yelowitz and I&nbsp;concluded that the number of working‐​age New York state residents on Medicaid who have incomes above the eligibility threshold rose by more than 80 percent between 2012 and 2017. We estimated that between 337,000 and 433,000 working‐​age New York state residents with income above the allowed limit are improperly enrolled in Medicaid.</p> </div> , <aside class="aside--right aside pb-lg-0 pt-lg-2"> <div class="pullquote pullquote--default"> <div class="pullquote__content h2"> <p>As a&nbsp;result of ObamaCare’s more generous Medicaid funding, many states — including New York — have stopped properly assessing whether applicants are eligible before they enroll.</p> </div> </div> </aside> , <div class="text-default"> <p>And nearly half of this improper enrollment is in New York City, with 30 percent in The Bronx and Queens, where a&nbsp;few neighborhoods have among the highest percentage of improper enrollees of anywhere in the country.</p> <p>In The Bronx, particularly the Concourse, Highbridge and Mount Eden regions, we found that roughly 40 percent of all working‐​age adults with incomes exceeding income eligibility thresholds were enrolled in Medicaid in 2017. The next‐​worst area is in Queens — the Elmhurst/​South Corona, Jackson Heights/​North Corona and Sunnyside/​Woodside regions. In those areas, there are likely tens of thousands of ineligible Medicaid enrollees.</p> <p>ObamaCare deserves much of the blame for the surge in improper enrollment. It created a&nbsp;new category of Medicaid recipients — lower‐​income, able‐​bodied, working‐​age adults — with the federal government paying a&nbsp;much larger share of their expenses than for traditional enrollees.</p> <p>From 2013 — the year before ObamaCare’s Medicaid expansion took effect — to 2018, there has been a&nbsp;surge of Medicaid payments out of compliance with legal criteria. In fact, improper Medicaid payments more than tripled.</p> <p>While states bear some of the burden for improper spending, most of the bill is picked up by the federal government. We estimated that improper payments now exceed 20 percent of federal Medicaid expenditures, an amount above $75 billion each year.</p> <p>As a&nbsp;result of ObamaCare’s more generous Medicaid funding, many states — including New York — have stopped properly assessing whether applicants are eligible before they enroll.</p> <p>While the health‐​care industry, particularly insurance companies, has benefitted from ObamaCare’s windfall of federal cash and improper Medicaid enrollment, traditional enrollees face a&nbsp;harder time obtaining care — and taxpayers are stuck with an enormous tab.</p> <p>The inspector general at the federal Department of Health and Human Services found substantial problems with New York state’s process for reviewing Medicaid eligibility. The state made large numbers of errors and did not always maintain documentation. An audit of the entire state’s program found 15 percent of applicants improperly enrolled. The size of the error was staggering, with the inspector general estimating that New York state improperly claimed more than $1.8 billion in a&nbsp;six‐​month period on behalf of more than 900,000 ineligible enrollees or people who were enrolled without having submitted all the proper documentation.</p> <p>In order to get a&nbsp;handle on its budget crisis, New York should conduct targeted eligibility reviews in The Bronx and Queens. If the state doesn’t act, the federal government must step in and require eligibility reviews in these hot spots and others around the country. Some level of government owes it to taxpayers and to those who are truly eligible to get enrollment right.</p> </div> Sat, 30 Nov 2019 09:54:00 -0500 Brian Blase, Aaron Yelowitz https://www.cato.org/publications/commentary/new-york-city-hot-spot-illegal-medicaid-enrollment Aaron Yelowitz’s co‐​written Wall Street Journal article, “Why Obama Stopped Auditing Medicaid,” is cited on KTSE’s Armstrong and Getty https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitzs-co-written-wall-street-journal-article-why-obama Tue, 26 Nov 2019 10:32:48 -0500 Aaron Yelowitz https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitzs-co-written-wall-street-journal-article-why-obama Medicaid Improper Payments are Much Worse Than Reported https://www.cato.org/blog/medicaid-improper-payments-are-much-worse-reported Aaron Yelowitz, Brian Blase <p>Earlier this week, Centers for Medicare &amp;&nbsp;Medicaid Services (CMS) <a href="https://www.cms.gov/newsroom/fact-sheets/2019-estimated-improper-payment-rates-centers-medicare-medicaid-services-cms-programs" target="_blank">raised</a> its estimate of Medicaid’s improper payments from $36 billion (9.8 percent of federal Medicaid expenditures) to $57 billion (14.9 percent of federal Medicaid expenditures). Actually, the situation is far worse than these estimates suggest. As we discussed in a&nbsp;Wall Street Journal <a href="https://www.wsj.com/articles/why-obama-stopped-auditing-medicaid-11574121931" target="_blank">op‐​ed</a> after the numbers were released, Medicaid’s improper payments now almost certainly exceed $75 billion – or more than 20 percent of federal Medicaid expenditures.</p> <p>This year’s report shows not only a&nbsp;significant increase in CMS’s estimate of improper payments. Its methodology also shows the agency has been hiding even larger improper payments for years. CMS estimates improper payments in the Medicaid program by auditing each state and DC once every three years and then using the most recent estimate available for each to construct a&nbsp;three‐​year rolling average. The 2018 report therefore covered fiscal years 2015–2017, while the 2019 report covered fiscal years 2016–2018.</p> <p>There’s one important caveat, however. The Obama administration <em>did not perform Medicaid eligibility audits for fiscal years 2014–2017</em>. Instead, it simply plugged the eligibility rate from the pre‐​Obamacare era into its improper payment calculations. This change would tend to hide any increases in improper enrollments that may have accompanied the implementation of ObamaCare’s Medicaid expansion. Indeed, other evidence suggests severe eligibility errors and problems accompanied the Medicaid expansion. As a&nbsp;result, the 2015, 2016, 2017, 2018, and 2019 reports likely underestimate the true extent of Medicaid improper payments because they use pre‐​ObamaCare data to describe what was happening under ObamaCare.</p> <p>So it’s a&nbsp;very big deal that CMS’s 2019 report showed a&nbsp;5.1 percentage point increase in the three‐​year moving average—from 9.8 percent to 14.9 percent—compared to the 2018 report. Note that while both reports’ averages use imputed improper payment rates for fiscal years 2016 and 2017 based on pre‐​2014 (i.e., pre‐​ObamaCare) data, the newest three‐​year average replaces the pre‐​ObamaCare rate used for 2015 with the actual, post‐​ObamaCare estimate for fiscal year 2018.</p> <p>Mathematically speaking, in order for the overall three‐​year average to rise by 5.1 percentage points, the improper payment rate estimate for FY 2018 must be 15.3 percentage points higher (5.1 x&nbsp;3&nbsp;years) than the FY 2015 estimate. Given that the true improper payment rate in 2015 was almost certainly at least 8&nbsp;percent, then the true improper payment rate in Medicaid (or more precisely, the real three‐​year rolling average from FYs 2016–2018) would then exceed 23 percent (15.3 percent + 8&nbsp;percent). In other words, as long as the improper payment rate in FY 2016 exceeded 5&nbsp;percent — and Medicaid’s improper payment rate has never been below 5&nbsp;percent — the true improper payment rate exceeds 20 percent.</p> <p>Even this much higher three‐​year rolling average is likely too low. CMS surveys the 50 states and DC in cycles – with 17 jurisdictions each year. The 2019 report updates the payment rate with “<a href="https://www.cms.gov/Research-Statistics-Data-and-Systems/Monitoring-Programs/Medicaid-and-CHIP-Compliance/PERM/Cycle_1" target="_blank">Cycle 1</a>” states – which includes Arkansas, Connecticut, Delaware, Idaho, Illinois, Kansas, Michigan, Minnesota, Missouri, New Mexico, North Dakota, Ohio, Oklahoma, Pennsylvania, Virginia, Wisconsin, and Wyoming. Recent OIG government audits of newly eligible Medicaid expansion enrollees known as “new adults” in <a href="https://oig.hhs.gov/oas/reports/region9/91602023.pdf" target="_blank">California</a>, <a href="https://oig.hhs.gov/oas/reports/region2/21501015.pdf" target="_blank">New York</a>, <a href="https://oig.hhs.gov/oas/reports/region7/71604228.pdf" target="_blank">Colorado</a>, and <a href="https://oig.hhs.gov/oas/reports/region4/41508044.pdf" target="_blank">Kentucky</a>, as well as non‐​newly eligible enrollees in <a href="https://oig.hhs.gov/oas/reports/region9/91702002.pdf" target="_blank">California</a>, <a href="https://oig.hhs.gov/oas/reports/region2/21601005.pdf" target="_blank">New York</a>, and <a href="https://oig.hhs.gov/oas/reports/region4/41608047.pdf" target="_blank">Kentucky</a> find serious program integrity issues and high improper payments during the 2014–2015 initial implementation of Obamacare. None of these states with well documented issues with their expansions were included in Cycle 1. Neither are <a href="http://app.lla.state.la.us/PublicReports.nsf/0/1CDD30D9C8286082862583400065E5F6/$FILE/0001ABC3.pdf" target="_blank">Louisiana</a> or <a href="https://sos.oregon.gov/audits/Documents/2017-25.pdf" target="_blank">Oregon</a>, where state audits showed significant problems with how those states were conducting eligibility reviews.</p> <p>In a&nbsp;forthcoming Mercatus paper, we find that the 12 expansion states with the largest rates of improper eligibility were New Mexico, California, Kentucky, Rhode Island, West Virginia, Oregon, Washington, Arkansas, Colorado, Louisiana, Montana, and New York. Yet, the CMS report only includes two of them (New Mexico and Arkansas). The result is that the 2019 estimate uses data that are not only not up to date, but also unrepresentative. An estimate employing up‐​to‐​date and representative data would show an even higher improper payment rate. Moreover, the sheer size and extent of improper enrollment problems in California will raise the cycle‐​specific rate for the 2020 report.</p> <p>Looking across the entire country and fully accounting for eligibility problems, it is possible, if not likely, that the improper payment rate is as high as 25 percent of federal Medicaid expenditures, or nearly $100 billion a&nbsp;year. That’s more than the entire $70 billion CMS actuaries <a href="https://www.cms.gov/Research-Statistics-Data-and-Systems/Research/ActuarialStudies/Downloads/MedicaidReport2017.pdf">estimate</a> the federal government spend on ObamaCare’s Medicaid expansion in 2018.</p> <p>Improper payments in Medicaid are a&nbsp;real problem for policymakers, and it is imperative that CMS finally take this problem seriously. Next week, the Mercatus Center will publish a&nbsp;new paper we authored that explains the problem in greater detail, including which states have the highest rates of improper payments, and makes a&nbsp;series of recommendations for reform.</p> Wed, 20 Nov 2019 16:32:22 -0500 Aaron Yelowitz, Brian Blase https://www.cato.org/blog/medicaid-improper-payments-are-much-worse-reported Why Obama Stopped Auditing Medicaid https://www.cato.org/publications/commentary/why-obama-stopped-auditing-medicaid Brian Blase, Aaron Yelowitz <div class="lead text-default"> <p>Medicaid expansion was a&nbsp;key component of ObamaCare. In 2014 when the expansion started, the feds stopped doing audits of states’ Medicaid eligibility determinations. The Obama administration’s goal was to build public support for the new law by signing up as many people as possible. Now, after a&nbsp;four‐​year hiatus, the Centers for Medicare and Medicaid Services have begun auditing program eligibility again. According to a&nbsp;report released Monday, the audits found “high levels of observed eligibility errors,” meaning a&nbsp;significant number of people are enrolled in Medicaid who shouldn’t be.</p> </div> , <div class="text-default"> <p>Our analysis of the CMS report suggests that the expansion appears to have more than tripled the amount of improper spending in the program. Twenty percent or more of Medicaid spending in 2019—an amount likely to exceed $75 billion—is improper. Before ObamaCare, the Medicaid improper‐​payment rate was 6%.</p> <p>Medicaid is a&nbsp;welfare program that finances health and long‐​term‐​care expenses of eligible recipients. America is a&nbsp;generous country, and it should have a&nbsp;public safety net to help individuals who confront extreme difficulties. Since welfare programs are expensive, it’s important that only eligible recipients receive benefits. Historically, the Medicaid program largely met that standard, covering lower‐​income children, pregnant women, adult caretakers, seniors and the disabled.</p> </div> , <aside class="aside--right aside pb-lg-0 pt-lg-2"> <div class="pullquote pullquote--default"> <div class="pullquote__content h2"> <p>The share of recipients who aren’t eligible has grown sharply since the expansion began in 2014.</p> </div> </div> </aside> , <div class="text-default"> <p>ObamaCare created a&nbsp;new category of Medicaid enrollees—working-age adults with income up to 138% of the poverty line, about $17,000 for a&nbsp;single person in 2019. For these new enrollees, states received a&nbsp;much higher reimbursement rate than for traditional Medicaid participants—100% from 2014 to 2016, gradually declining to 90% in 2020, where it is scheduled to remain.</p> <p>ObamaCare’s Medicaid expansion presented states with an opportunity to increase substantially the federal dollars flowing into their coffers. It also offered states the chance to game the program’s rules. As a&nbsp;result of the elevated reimbursement, states have a&nbsp;major incentive to classify people as expansion enrollees. Some who were eligible before ObamaCare were reclassified as expansion enrollees so states could take advantage of the higher reimbursement rate.</p> <p>Higher overall Medicaid payments came with benefits for state‐​level interest groups that profited from maximizing enrollment. Insurers have reaped substantial profits from the Medicaid expansion—owing in part to large government payments for people who are enrolled but don’t go to the doctor or use much medical care.</p> <p>Since states view the Medicaid expansion as a&nbsp;cash cow, they have generally failed to conduct proper eligibility reviews. One&nbsp;<a href="https://oig.hhs.gov/oas/reports/region9/91702002.pdf?mod=article_inline" target="_blank">federal audit</a> by the Health and Human Services Department’s inspector general found that more than half of sampled enrollees in California’s Medicaid program were either improperly enrolled or potentially improperly enrolled. Whether out of greed or incompetence, many states neglect to obtain proper documentation and fail to verify income eligibility and citizenship.</p> <p>New research we have conducted shows the problem is common in expansion states, but most severe in California, Kentucky, New Mexico, New York and West Virginia. Overall, we estimate that between 2.3 million and 3.3 million people with income above eligibility thresholds—and who would not be eligible for Medicaid for another reason like pregnancy or disability—are enrolled in Medicaid in expansion states. Within these states, there are some areas, such as New York City and Los Angeles, where the problem appears so large that it suggests purposeful and fraudulent abuse on the part of local officials and the medical industry.</p> <p>ObamaCare created an incentive for states to view the Medicaid expansion population as a&nbsp;cash cow, but CMS deserves part of the blame. In addition to canceling eligibility audits for four years, the agency has never taken meaningful actions to minimize improper payments from the expansion. With limited federal oversight and little if any effective federal action to penalize states for improper eligibility determinations, states have almost no incentive to administer their programs responsibly or lawfully.</p> <p>The findings in the CMS report confirm that Medicaid’s improper‐​spending problem is large and growing. Federal policy makers have a&nbsp;responsibility to those who are truly eligible and most need Medicaid, as well as to the nation’s taxpayers, to address the problem. They should start by requiring eligibility redeterminations in areas where the problem is most severe and by recouping funds improperly claimed by states.</p> </div> Wed, 20 Nov 2019 09:07:26 -0500 Brian Blase, Aaron Yelowitz https://www.cato.org/publications/commentary/why-obama-stopped-auditing-medicaid FamilyTreeDNA, Government Overreach, and Unethical Nudges https://www.cato.org/blog/familytreedna-government-overreach-unethical-nudges Aaron Yelowitz <p>A disturbing story about <a href="https://www.wsj.com/articles/customers-handed-over-their-dna-the-company-let-the-fbi-take-a-look-11566491162">FamilyTreeDNA</a> highlights issues about consumer privacy, government collaboration, and poor stewardship by a&nbsp;private company. Digging deeper, the story also highlights how behavioral economics can go awry, through self‐​serving choices by a&nbsp;moralistic CEO that violate basic ethical principles of choice architecture design. Bad nudges is an issue I&nbsp;have highlighted before in the context of <a href="https://www.cato.org/blog/bad-nudges-kentucky-medicaid">Kentucky Medicaid</a> plan choice and <a href="https://www.cato.org/blog/government-mandated-state-run-auto-iras-can-cause-real-harm">state‐​run auto‐​IRAs</a>. <br><br /> <br> The short version: FamilyTreeDNA’s database contains more than 1.5 million customers, and the FBI approached company president Bennett Greenspan in late 2017 and early 2018 to access those records in hopes of finding genetic links for some unsolved crimes. As the Wall Street Journal notes: <br><br /> <br><em>He didn’t tell the FBI attorney to come back with a&nbsp;court order. He didn’t stop to ponder the moral quandaries. He said yes on the spot. “I have been a&nbsp;CEO for a&nbsp;long time,” said Mr. Greenspan, 67&nbsp;years old, who founded the Houston‐​based company in 1999. “I have made decisions on my own for a&nbsp;long time. In this case, it was easy. We were talking about horrendous crimes. So I&nbsp;made the decision.”</em> <br><br /> <br> Any libertarian would certainly agree that consumers and companies should be free to come to any agreement they want on sacrificing personal privacy for other product characteristics (including lower prices). Even with an open‐​ended user agreement, it is hard to fathom that even the most imaginative users from 15&nbsp;years ago would have envisioned the sort of law enforcement overreach that we see today. If informed, some subset of customers would likely support FamilyTreeDNA’s collaboration with the FBI. The user agreement did not require the company to inform customers that the FBI was searching their records, and the company did not inform customers until after <a href="https://www.buzzfeednews.com/article/salvadorhernandez/family-tree-dna-fbi-investigative-genealogy-privacy">Buzzfeed</a> revealed the collaboration in January 2019. <br><br /> <br> Although the CEO appears to be an enthusiastic participant in the FBI’s dragnet, this may be the exception, rather than the rule. Other DNA testing companies – such as 23andMe, Ancestry, and MyHeritage – do not collaborate with law enforcement unless legally required to do so. One must wonder how much extra legal costs are borne by private companies from law enforcement overreach like this, and how much it would cost a&nbsp;company to vigorously fight back against the fishing expeditions? Surely, the cost of law enforcement overreach is passed on to customers who pay more in submission fees, in order to have their privacy invaded. <br><br /> <br> It is important to put the company’s subsequent response to the fallout into a&nbsp;behavioral economic lens. <br><br /> <br><em>In March (2019), FamilyTreeDNA said it figured out a&nbsp;way <strong>to allow customers to opt out of law‐​enforcement matching</strong> but still see if they matched with regular customers. … (Mr. Greenspan) said <strong>less than 2% of customers have requested opting out</strong> of law‐​enforcement searches.</em> <br><br /> <br> In his pioneering work, Prof. Cass Sunstein lays out <a href="http://www.law.harvard.edu/programs/olin_center/papers/pdf/Sunstein_809.pdf">ethical considerations</a> for choice architecture. He argues that the objective of nudging is to “influence choices in a&nbsp;way that will make the choosers better off, as judged by themselves.” In this context, when confronted with obvious outrage and bad publicity, FamilyTreeDNA had important decisions to make. Sunstein’s “as judged by themselves” principle would suggest the opposite choice architecture: <strong>the company should have set the default as automatic opt‐​out of law enforcement matching,</strong> and allowed users to opt‐​in to law enforcement matching if they so decided. Many of the 1.5 million customers are likely infrequent, inactive users of the website, and many were likely unaware of the FBI collaboration, even after the news broke. They would be appalled by the collaboration. Mr. Greenspan’s opt‐​out figure of 2% strikes me as a&nbsp;very large response, given that FamilyTreeDNA has customers going back 20&nbsp;years, and many likely ignore emails and news stories about this scandal. <br><br /> <br> The criticism of this company’s choice architecture – and feature stories in prominent newspapers – of course would not exist without unabated government overreach. From a&nbsp;handful of inquiries in early 2018, there are now 50 law‐​enforcement agencies requesting matching from FamilyTreeDNA. Buyer beware.</p> Sun, 25 Aug 2019 15:30:00 -0400 Aaron Yelowitz https://www.cato.org/blog/familytreedna-government-overreach-unethical-nudges ObamaCare’s Medicaid Deception https://www.cato.org/publications/commentary/obamacare-medicaid-deception Brian Blase, Aaron Yelowitz <div class="lead text-default"> <p>ObamaCare wasn’t supposed to give free health insurance to everybody. The Affordable Care Act’s authors expected the poor would enroll in Medicaid, while those with higher incomes would buy coverage through the new insurance exchanges, with subsidies that decrease as income rises.</p> </div> , <div class="text-default"> <p>It isn’t working. A&nbsp;<a href="https://www.nber.org/papers/w26145?utm_campaign=ntwh&amp;utm_medium=email&amp;utm_source=ntwg1&amp;mod=article_inline" target="_blank">study</a>&nbsp;published this week by the National Bureau of Economic Research finds that in several Medicaid‐​expansion states most people who gained coverage have enrolled in Medicaid regardless of their income. In practice, ObamaCare has turned Medicaid into an entitlement program for the middle class.</p> <p>Using data from U.S. Census Bureau’s American Community Survey, the authors assessed coverage changes from 2012–17&nbsp;in nine states that expanded Medicaid vs. 12 states that didn’t. They uncovered a&nbsp;huge problem. In 2017 alone, in those nine states, “around 800,000 individuals … appeared to gain Medicaid coverage for which they were seemingly income‐​ineligible.”</p> <p>ObamaCare is supposed to make Medicaid available to households with incomes below 138% of the poverty line, or nearly $36,000 for a&nbsp;family of four. In the nine states—Arkansas, Kentucky, Michigan, Nevada, New Hampshire, New Mexico, North Dakota, Ohio and West Virginia—the authors found that among households with incomes 138% to 250% of the poverty line (about $65,000 for a&nbsp;family of four), some 78% that gained coverage had improperly enrolled in Medicaid. That was also true of 65% of the population above 250% of poverty that gained coverage.</p> <p>This isn’t a&nbsp;matter of growing pains. Improper enrollment has increased over time. It was two to three times as prevalent in 2017 as in 2014. It’s a&nbsp;systematic problem with ObamaCare in practice.</p> <p>These estimates likely understate the true problem. People tend to minimize total income when responding to surveys. The authors chose these nine states because they adopted the ObamaCare expansion in 2014 and didn’t previously cover any able‐​bodied, working‐​age people in Medicaid. The nine account for less than 20% of the total population living in expansion states.</p> <p>There’s evidence of massive improper enrollment in other states. According to 2018 reports by the Inspector General’s Office at the Department of Health and Human Services, 25% of Medicaid expansion enrollees were likely ineligible in both California and New York.</p> <p>A state audit in Louisiana found 82% of expansion enrollees were ineligible at some point during the year they were enrolled. The central problem appears to be the state’s reliance on the federal exchange website to determine eligibility. People who entered no income simply to explore their options were automatically enrolled in Medicaid. Eligibility works the same way in another seven states.</p> <p>The number of ineligible enrollees in these three states alone almost certainly exceeds one million people. These findings should alarm Americans across the political spectrum. They show that complicated government programs often bear little resemblance to planners’ designs. ObamaCare has turned out to be a&nbsp;giant welfare program, with millions of working‐ and middle‐​class Americans improperly receiving Medicaid—a reflection of the unpopularity of the exchange policies and incompetence of government oversight.</p> <p>States that opted not to expand Medicaid have been much better able to preserve private coverage. Employer‐​sponsored coverage has steadily grown in nonexpansion states with virtually no growth in expansion states.</p> <p>The Centers for Medicare and Medicaid Services need to do much more. While CMS cannot undo the structural flaws at the core of ObamaCare, they can use their oversight and enforcement powers to minimize the massive improper and fraudulent expansion enrollment. Medicaid needs to be protected and taxpayer dollars preserved for the disabled and low‐​income children, pregnant women and seniors.</p> </div> Wed, 14 Aug 2019 14:42:26 -0400 Brian Blase, Aaron Yelowitz https://www.cato.org/publications/commentary/obamacare-medicaid-deception Government Mandated, State‐​Run Auto‐​IRAs Can Cause Real Harm https://www.cato.org/blog/government-mandated-state-run-auto-iras-can-cause-real-harm Aaron Yelowitz <p>A number of states have recently enacted employer mandates that force companies who don’t offer retirement plans to enroll their workers in a&nbsp;state‐​run, auto‐​IRA plan. Oregon’s program – known as OregonSaves – is the oldest and most established. By mid‐​2020, <a href="https://www.oregon.gov/retire/SiteAssets/Pages/Newsroom/2018%20OregonSaves%20Annual%20Report%20FINAL.pdf">Oregon’s</a> mandate will cover all companies; it currently covers companies with twenty or more workers. <br><br /> <br> One myth – perpetuated by the <a href="https://www.nelp.org/blog/claims-auto-iras-will-backfire-dont-hold-water/">National Employment Law Project</a> – is that state mandates expand opportunity to retirement savings, especially for low‐​income workers. They don’t. OregonSaves initially defaults worker contributions into a&nbsp;conservative <a href="https://www.ssga.com/cash/funds/SSIXX/fund_overview_SSIXX.html">capital preservation fund</a> before redirecting contributions to a&nbsp;life‐​cycle fund once balances exceed $1,000. Since inception in 2004, the capital preservation fund has offered a&nbsp;paltry nominal return of 1.52% (essentially an inflation‐​adjusted return of 0%). OregonSaves also assesses an administrative fee of <a href="https://saver.oregonsaves.com/home/program-details.html">100 basis points</a> (that is, 1%) regardless of investment choices, further diminishing this return. This set‐​up isn’t really an opportunity for Oregon workers, because they already have access to Roth IRAs and investments with a&nbsp;more beneficial set‐​up. A&nbsp;25‐​year‐​old worker might actively choose <a href="https://fundresearch.fidelity.com/mutual-funds/fees-and-prices/315793851">a&nbsp;life cycle fund</a> with no minimums for initial investment or additional contributions, along with administrative fees of 75 basis points, significantly lower than OregonSaves. Choosing an index fund that tracks the <a href="https://fundresearch.fidelity.com/mutual-funds/summary/315911750">S&amp;P 500</a> could have administrative fees as low as 1.5 basis points. Without mandating Oregon employers to enroll their workers, OregonSaves would struggle to compete in a&nbsp;vibrant marketplace with many inexpensive alternatives for retirement contributions. <br><br /> <br> If government mandates don’t improve access to retirement plans, why have the program? The real reason is that the programs increase participation through inertia; simply put, many workers are asleep at the wheel. Many workers don’t take active steps to plan for retirement regardless of how a&nbsp;program is designed. If the default choice is to actively enroll, many workers won’t participate. If the default choice is automatic enrollment with an opt‐​out option, many workers do participate. Oregon’s 28% opt‐​out rate is relatively high, highlighting some of the problems of the program’s design. Among those enrolled, fully 93% of participants stick with the specified contribution rate and an astonishing 79% of all fund balances are invested in the capital preservation fund. Almost all remaining balances are invested in target date funds, likely for workers who have exceeded the $1,000 contribution. <br><br /> <br> Worker inertia is real, meaning that design choices like opting in or out, asset classes and contribution rates are likely to stick. The one‐​size‐​fits‐​all design of OregonSaves can cause real harm for many workers, an issue I&nbsp;explored with my colleagues in a&nbsp;new study for <a href="https://jor.iijournals.com/content/6/2/27">Journal of Retirement</a>. If OregonSaves were adopted nationally, 24.2 million workers aged 25–64 would initially be opted‐​in. Approximately 33% of affected workers carry high‐​interest credit card debt, with balances averaging nearly $5,500. Around 15% of affected workers struggle to meet basic needs like paying rent or utility bills. Workers in these situations come out ahead by paying down debt or meeting basic needs, and siphoning off 5% of their paycheck will likely worsen their overall financial situation. <br><br /> <br> Financial planning websites consistently emphasize paying off revolving high‐​interest debt before saving for retirement (unless a&nbsp;company offers a&nbsp;match rate), yet auto‐​IRAs fail to take these investment lessons into account. <a href="https://www.nelp.org/blog/claims-auto-iras-will-backfire-dont-hold-water/">Advocates for government mandates</a> emphasize the benefits of compounding for assets in an IRA, while curiously ignoring the reality that unpaid debt compounds in the exact same manner! At an 18% interest rate, an unpaid $5,500 credit card debt would mushroom to $28,800&nbsp;in ten years. The same amount of money directed towards OregonSaves might accumulate to $12,900 under rosy assumptions about investment returns. Ultimately, our study shows a&nbsp;significant number of workers are in situations like this, and auto‐​IRAs would do more harm than good for them.</p> Fri, 22 Mar 2019 12:58:00 -0400 Aaron Yelowitz https://www.cato.org/blog/government-mandated-state-run-auto-iras-can-cause-real-harm Early Effects of the Affordable Care Act on Health Care Access, Risky Health Behaviors, and Self‐​Assessed Health https://www.cato.org/publications/research-briefs-economic-policy/early-effects-affordable-care-act-health-care-access Charles Courtemanche, James Marton, Benjamin Ukert, Aaron Yelowitz, Daniela Zapata <div class="lead text-default"> <p>The goal of the Patient Protection and Affordable Care Act (ACA) was to achieve nearly universal health insurance coverage in the United States through a&nbsp;combination of policies largely implemented in 2014. Several recent studies have shown that the ACA led to gains in insurance coverage. We evaluate whether or not such coverage increases translated to changes in access to care, risky health behaviors, and, ultimately, short‐​run health outcomes.</p> </div> , <div class="text-default"> <p>A number of 2014 ACA provisions involved overhauling nongroup insurance markets in an effort to ensure that one’s health history did not provide a&nbsp;barrier to obtaining coverage. Specific regulations included guaranteed issue laws, which forbid insurers from denying coverage on the basis of an applicant’s health status, and modified community rating, which imposes uniform premiums regardless of observable applicant characteristics aside from age and smoking status. In addition, the federal government established a&nbsp;health insurance marketplace to facilitate insurance purchases for individuals and small businesses. Each state was given the option of establishing its own insurance marketplace, and 15 did so in 2014.</p> <p>These reforms alone would likely lead to an adverseselection death spiral, with the influx of high‐​cost beneficiaries causing relatively low‐​cost beneficiaries to drop their coverage, thus driving up premiums for those remaining in the insurance pool. This concern motivated another component of the ACA: the individual mandate. Beginning in 2014, individuals deemed to be able to afford coverage but electing to remain uncovered were penalized. The largest penalty that could be imposed was the maximum of either the total annual premium for the national average price of a&nbsp;bronze exchange plan, or $285 and $975&nbsp;in 2014 and 2015, respectively. In addition, an employer mandate, which required employers with 100 or more full‐​time equivalent employees to offer “affordable” coverage to at least 95 percent of their full‐​time employees and their dependents (children up to age 26) or face a&nbsp;penalty, took effect in 2015.</p> <p>The remaining challenge associated with promoting universal coverage—affordability—was addressed by the ACA in 2014&nbsp;in two ways. First, sliding scale subsidies in the form of premium tax credits became available to consumers in every state with incomes of 100 percent to 400 percent of the federal poverty level (FPL) who did not qualify for other affordable coverage. Second, in states that opted to expand Medicaid via the ACA, low‐​income adults (with incomes at or below 138 percent of the FPL) who were not elderly, disabled, or parents of a&nbsp;dependent child became eligible for Medicaid coverage. Previously, Medicaid eligibility was typically restricted to those with low incomes among specific groups (categories of eligibility), such as children, single parents, pregnant women, the disabled, and the elderly. According to the Kaiser Family Foundation, 27 states participated in the Medicaid expansion in 2014, with three more implementing it in 2015 and another two in 2016.</p> <p>Theoretically, the expansion of insurance coverage brought about by the ACA should increase access to care because of the reduction in out‐​of‐​pocket costs, but this is not automatically the case. On the demand side, newly insured individuals may not have sufficient knowledge of the health care system to easily secure a&nbsp;regular primary care doctor. Stephen Somers and Roopa Mahadevan of the Center for Health Care Strategies report that only 12 percent of adults have proficient health literacy. On the supply side, concerns have been raised about whether there are sufficient numbers of primary care physicians to treat all of these newly insured patients. While the federal government increased Medicaid primary care reimbursement rates to Medicare levels in 2013 and 2014, only a&nbsp;few states fully maintained this “fee bump” in 2015.</p> <p>Insurance coverage expansions could influence risky health behaviors—such as smoking, drinking, and overeating—in either direction. On the one hand, improved access to care among the affected population could translate to improvements in health behaviors via information, accountability, or treatments such as smoking‐​cessation drugs or weight‐​loss programs. Conversely, insurance expansions can theoretically worsen health outcomes through ex ante moral hazard, as the reduction in financial risks associated with unhealthy behaviors incentivizes such behaviors. Moreover, income effects from gaining free or subsidized coverage could influence behaviors by enabling consumers to spend money they had budgeted for the direct purchase of health care on alcohol, cigarettes, and junk food or, conversely, on healthy food and gym memberships.</p> <p>The net effect of insurance expansions on population health depends on the changes in both access to care and health behaviors and, therefore, is also theoretically ambiguous. The extent to which increased health care utilization translates to better population health depends on the distribution of affected individuals’ initial locations along the health production function. Evidence suggests that “flat of the curve” care—perhaps due to uncertainty over treatment effectiveness, the principal‐​agent nature of the patient‐​doctor relationship, fee‐​for‐​service reimbursement, lack of coordination across health care providers, or malpractice liability—is common in the United States. Moreover, the same issues with health literacy that could hamper efforts by the newly insured to find a&nbsp;primary care doctor could also limit their ability to understand and comply with treatment recommendations.</p> <p>The purpose of our research is to estimate the impact of the ACA’s 2014 provisions on a&nbsp;variety of outcomes related to health care access, risky health behaviors, and self‐​assessed health. We separately identify the effects of the private and Medicaid‐​expansion portions of the ACA by estimating the impact of the ACA on insurance coverage using differences across local areas in pretreatment uninsured rates. To be more specific, we use the differences coming from time, state Medicaid‐​expansion status, and local area pretreatment uninsured rates. If our objective were merely to isolate the effect of the Medicaid expansion, that could potentially be achieved with a&nbsp;simpler model comparing changes in states that expanded Medicaid to changes in nonexpansion states. However, identifying the impact of the other components of the ACA (e.g., mandates, subsidies, and marketplaces) is more difficult due to their national nature. We therefore exploit an additional layer of plausibly exogenous variation arising from the fact that universal coverage initiatives provide the most intense treatments in areas with high uninsured rates.</p> <p>Our data come from the 2011–2015 waves of the Behavioral Risk Factor Surveillance System (BRFSS), with the sample restricted to nonelderly adults. The BRFSS is well suited to our study for three reasons. First, it includes a&nbsp;wide range of questions on health care access and self‐​assessed health. Second, with over 300,000 observations per year, it is large enough to precisely estimate the effects of state‐​level interventions. Third, it was among the first large‐​scale health data sets to release data from 2015, allowing us to examine two calendar years of data after the full implementation of the ACA.</p> <p>Our results suggest that the ACA substantially improved access to health care among nonelderly adults. Gains in insurance coverage were 8.3 percentage points in Medicaidexpansion states compared to 5.3 percentage points in nonexpansion states, while reductions in cost being a&nbsp;barrier to care were 5.1 percentage points in expansion states and 2.6 percentage points in nonexpansion states. The ACA also increased the probabilities of having a&nbsp;primary care doctor and a&nbsp;checkup by 3.0 and 2.4 percentage points, respectively, in non‐​Medicaid‐​expansion states, with the effects not being statistically different in expansion states. Gains in access were generally largest among individuals with lower incomes.</p> <p>However, the effects of the ACA on risky health behaviors and self‐​assessed health were less pronounced—at least after two years. For risky behaviors, we examine smoking, alcohol consumption, and body mass index. For health outcomes, we examine self‐​assessed health, days in poor mental health, days in poor physical health, and days with health‐​related functional limitations. For the full sample, we find no statistically significant impacts on any of the risky behavior or health outcomes in either Medicaid‐​expansion or nonexpansion states. This general pattern of null results persists even among the lower‐​income subsample, though we do observe a&nbsp;marginally significant improvement in mental health in Medicaidexpansion states for that group.</p> <p>In summary, the insurance expansions in 2014&nbsp;in the ACA had a&nbsp;large impact on insurance coverage, along with a&nbsp;large price tag. One of the key stated benefits for expanding insurance coverage is to improve the overall health of the population; our work shows that at least in the short run, such benefits were minimal.</p> <p><strong>NOTE</strong>: <br>This research brief is based on Charles Courtemanche, James Marton, Benjamin Ukert, Aaron Yelowitz, and Daniela Zapata, “Early Effects of the Affordable Care Act on Health Care Access, Risky Health Behaviors, and Self‐​Assessed Health,” <em>Southern Economic Journal</em> 84, no. 3 (January 2018): 660–91.</p> </div> Wed, 11 Apr 2018 00:00:00 -0400 Charles Courtemanche, James Marton, Benjamin Ukert, Aaron Yelowitz, Daniela Zapata https://www.cato.org/publications/research-briefs-economic-policy/early-effects-affordable-care-act-health-care-access Bad Nudges — Kentucky Medicaid https://www.cato.org/blog/bad-nudges-kentucky-medicaid Aaron Yelowitz <p>In their highly influential book describing behavioral economics, <em>Nudge</em>, Richard H. Thaler and Cass R. Sustein devote 2 pages to the notion of "bad nudges." They describe a "nudge" as any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives. The classic example of a nudge is the decision of an employer to "opt-in" or "opt-out" employees from a 401(k) plan while allowing the employee to reverse that choice; the empirical evidence strongly suggests that opting employees into such plans dramatically raises 401(k) participation. Many parts of the book advocate for more deliberate choice architecture on the part of the government in order to "nudge" individuals in the social planner's preferred direction.&#13;<br /> &#13;<br /> Thaler and Sunstein provide short discussion and uncompelling examples of bad nudges. They correctly note "In offering supposedly helpful nudges, choice architects may have their own agendas. Those who favor one default rule over another may do so because their own economic interests are at stake." (p. 239) With respect to nudges by the government, their view is "One question is whether we should worry even more about public choice architects than private choice architects. Maybe so, but we worry about both. On the face of it, it is odd to say that the public architects are always more dangerous than the private ones. After all, managers in the public sector have to answer to voters, and managers in the private sector have as their mandate the job of maximizing profits and share prices, not consumer welfare."&#13;<br /> &#13;<br /> In <a href="https://doi.org/10.1016/j.jhealeco.2017.04.006">my recent work</a> (with Jim Marton and Jeff Talbert), we show how bad nudges by public officials can work in practice through a compelling example from Kentucky. In 2012, Kentucky implemented Medicaid managed care statewide, auto-assigned enrollees to three plans, and allowed switching. This fits in with the "choice architecture" and "nudge" design described by Thaler and Sunstein. One of the three plans – called KY Spirit – was decidedly lower quality than the other two plans, especially in eastern Kentucky. For example, KY Spirit was not able to contract with the dominant health care provider in eastern Kentucky due to unsuccessful rate negotiations. KY Spirit’s difficulties in eastern Kentucky were widely reported in the press, so we would expect there to be greater awareness of differences in MCO provider network quality in that region.&#13;<br /> &#13;</p> <p>Given the virtually identical and non-existent financial differences across the three Medicaid plans (they were essentially free to Medicaid clients), the standard economic framework with rational consumers and trivial transaction costs would predict all enrollees would switch out of lower quality plans. In this case, it would suggest mass defections from KY Spirit. In contrast, the “nudge” framework suggests enrollees would be for more likely to remain in inferior plans. The nudge – in this case a bad nudge – worked. In each of the other two plans – both of higher quality – approximately 95% of those assigned to those plans stayed in them. For KY Spirit, the percentage was lower, but very far from the prediction of full-scale exit. Specifically, 57% of those assigned to KY Spirit remained enrolled in the plan in 2012, despite its well-documented problems. For sicker individuals, 44% remained in KY Spirit, despite the serious problems in accessing healthcare providers. Very few individuals who opted out of their assigned health plan made the active choice to enroll in KY Spirit, consistent with the notion of its low quality. Of more than 37,000 individuals in eastern Kentucky assigned to the other two health plans, slightly more than 100 actively moved into KY Spirit.&#13;<br /> &#13;<br /> Why would public officials assign Medicaid enrollees to a low quality health care plan? After all, virtually all examples of government nudges in the Thaler and Sunstein book portray officials as steering clients in the right direction. In the Kentucky context, the underlying motivation appears to be program costs. The state paid different reimbursement rates to each of the three health plans, and most of the time, KY Spirit was the "low cost, low quality” plan. In reality, this “bad nudge” – from the Medicaid enrollee’s perspective – was a cost saving from the taxpayer’s point of view. Compare to an objective of maximizing the quality of plans for Medicaid enrollees, the actual plan assignment which included some “bad nudges” reduced program costs by approximately 5%.&#13;<br /> &#13;<br /> Although policymakers might be applauded in this case for reigning in program costs through behavioral economics, it is far from the optimistic framework portrayed in Thaler and Sunstein about maximizing client interest.</p> <p></p> Tue, 27 Mar 2018 10:32:00 -0400 Aaron Yelowitz https://www.cato.org/blog/bad-nudges-kentucky-medicaid Consequences (and Repeal) of the Affordable Care Act https://www.cato.org/multimedia/cato-daily-podcast/consequences-repeal-affordable-care-act Aaron Yelowitz, Caleb O. Brown <p>What has the Affordable Care Act meant for health insurance coverage? What should repeal look like? Aaron Yelowitz comments.</p> Tue, 28 Feb 2017 20:51:00 -0500 Aaron Yelowitz, Caleb O. Brown https://www.cato.org/multimedia/cato-daily-podcast/consequences-repeal-affordable-care-act Aaron Yelowitz discusses his Policy Analysis, “Menu Mandates and Obesity: A Futile Effort,” on WJIM’s The Steve Gruber Show https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitz-discusses-policy-analysis-menu-mandates-obesity Mon, 25 Apr 2016 10:49:00 -0400 Aaron Yelowitz https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitz-discusses-policy-analysis-menu-mandates-obesity Aaron Yelowitz’s Policy Analysis, “Menu Mandates and Obesity: A Futile Effort,” is cited on KTRH Radio https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitzs-policy-analysis-menu-mandates-obesity-futile Thu, 14 Apr 2016 13:42:00 -0400 Aaron Yelowitz https://www.cato.org/multimedia/media-highlights-radio/aaron-yelowitzs-policy-analysis-menu-mandates-obesity-futile Menu Mandates and Obesity: A Futile Effort https://www.cato.org/publications/policy-analysis/menu-mandates-obesity-futile-effort Aaron Yelowitz <div class="lead text-default"> <p>One provision of the Patient Protection and Affordable Care Act (ACA) that has been delayed until 2017 is a&nbsp;federal mandate for standard menu items in restaurants and some other venues to contain nutrition labeling. The motivation for so‐​called “menu mandates” is a&nbsp;concern about rising obesity levels driven largely by Americans’ eating habits. Menu mandates have been implemented at the state and local level within the past decade, allowing for a&nbsp;direct examination of the short‐​run and long‐​run effects on outcomes such as body mass index (BMI) and obesity. Drawing on nearly 300,000 respondents from the Behavioral Risk Factor Surveillance System (BRFSS) from 30 large cities between 2003 and 2012, we explore the effects of menu mandates. We find that the impact of such labeling requirements on BMI, obesity, and other health‐​related outcomes is trivial, and, to the extent it exists, it fades out rapidly. For example, menu mandates would reduce the weight of a&nbsp;5′10″ male adult from 190&nbsp;pounds to 189.5&nbsp;pounds. For virtually all groups explored, the long‐​run impact on body weight is essentially zero. Analysis of subgroups suggests that to the extent that menu mandates affect short‐​run outcomes, they do so through a “novelty effect” that wears off quickly. Subgroups that were thought likely to experience the largest gains in knowledge from such mandates exhibit no short‐​run or long‐​run changes in weight.</p> </div> , <div class="text-default"> <h2>Menu Mandates and Obesity</h2> <h2><a id="sec-2"></a><strong>Introduction</strong></h2> <p>The prevalence of obesity has increased markedly in the United States over time and has affected all socioeconomic groups.<sup><a href="#cite-1">1</a></sup>, <sup><a href="#cite-2">2</a></sup>, <sup><a href="#cite-3">3</a></sup> Although the estimated cost of obesity—in terms of disease, medical visits, lost work days, and other outcomes—varies widely, some have argued that these costs represent a rationale for government intervention to reduce obesity‐​related externalities.<sup><a href="#cite-4">4</a></sup>, <sup><a href="#cite-5">5</a></sup></p> <p>The Patient Protection and Affordable Care Act is the most significant government overhaul of the U.S. healthcare system since the passage of Medicare and Medicaid in the 1960s. One often overlooked provision, Section 4205, mandates that calorie information be provided on menus of restaurants and numerous other venues.<sup><a href="#cite-6">6</a></sup> When fully implemented, this “menu mandate” will affect 300,000 establishments, and the breadth of the Food and Drug Administration’s (FDA) final rule surprised even health advocates.<sup><a href="#cite-7">7</a></sup>, <sup><a href="#cite-8">8</a></sup> Chain restaurants, movie theaters, grocery stores (for their salad or hot bars), and vending machines will be forced to provide calorie counts. Although this FDA regulation was supposed to be effective in December 2015, it was pushed back and will now be implemented in 2017.<sup><a href="#cite-9">9</a></sup></p> <p>The stated motivation for such menu mandates is to reduce the number of overweight and obese Americans by reducing their consumption of calories. A significant portion of food expense and calories comes from foods prepared outside the home, and government officials believe that many people do not know (and may underestimate) the caloric content of such food.<sup><a href="#cite-10">10</a></sup> The federal mandate was preceded by similar efforts at the state and local levels within the past decade, perhaps the best known of which was New York City’s menu mandate in 2008. At the time, some argued that menu mandates could lead to substantial reductions in weight—roughly 7.5 pounds per year, or 106 calories per fast‐​food transaction.<sup><a href="#cite-11">11</a></sup></p> <p>To date, the most convincing evidence concerning the effects of menu mandates—both in New York City and elsewhere—has been on calorie consumption associated with individual transactions. The evidence on the broader effectiveness of such mandates is mixed; we discuss it later. However, more important than any one transaction is whether menu mandates have any long‐​lasting impact on body weight or obesity. The principal contribution of this analysis is to explore this issue by using publicly available data on nearly 300,000 respondents from the Behavioral Risk Factor Surveillance System for 30 large cities between 2003 and 2012. On a staggered basis, some of these cities implemented menu mandates, while others did not. This paper finds that the impact of such mandates on body weight is trivial, and to the extent an impact exists, it fades out rapidly. For example, menu mandates would reduce the weight of a 5′10″ male adult from 190 pounds to 189.5 pounds. For virtually all groups explored, the long‐​run impact on body weight of menu mandates is essentially zero. This evidence demonstrates the futility of government efforts at altering individuals’ preferences regarding the food they eat; the lack of benefit, in conjunction with costs both to consumers and businesses, shows that government‐​imposed menu mandates are ill‐​advised.</p> <h2><a id="sec-2.1"></a>The Benefits of Menu Mandates</h2> <p>Bollinger et al. discuss the potential impact of menu mandates.<sup><a href="#cite-12">12</a></sup> Learning information about calories contained in food and beverages may lead to healthier purchases by consumers at chain restaurants. However, customers may care mostly about convenience, price, and taste, with calories being relatively unimportant. It may also be the case that those who do care about calories are already well‐​informed; such nutrition information is available for the motivated customer.</p> <p>How menu mandates affect behavior in the long run, or outside of the restaurant setting, is less clear. With respect to long‐​run behavior, such mandates may improve a customer’s knowledge of calories (a “learning effect”) or sensitivity to calories (a “salience effect”).<sup><a href="#cite-13">13</a></sup> To the extent that menu mandates improve learning and correct misperceptions about food calories, the effects of menu mandates are more likely to be permanent. To the extent they simply make calories more salient, the effects are more likely to be short‐​lived. Efforts to make unwanted information more salient—from web banners to graphic tobacco warning labels—tend to be ineffective, especially after the initial novelty wears off.<sup><a href="#cite-14">14</a></sup> Cantor et al. surveyed consumers in New York City immediately after the menu mandate took effect in 2008, and at three points during 2013–2014.<sup><a href="#cite-15">15</a></sup> They found that the percentage of respondents noticing and using the information declined in each successive period, and that there were no statistically significant changes in calorie levels or visits to fast‐​food restaurants.</p> <p>In addition to these responses, it is also possible that restaurants innovate by offering more low‐​calorie items in the long run, making the mandate more impactful. Outside the restaurant setting, consumers’ exposure to calorie information may make them generally more aware and attentive to the nutritional value of the foods they eat.<sup><a href="#cite-16">16</a></sup> On the other hand, people may offset changes in their calorie consumption at restaurants by changing what they eat at home.</p> <p>Ultimately, the evidence on the effects of menu mandates on caloric intake at restaurants is mixed. Studying the implementation of the menu mandate in New York City, Bollinger et al. find that average calories consumed per transaction at Starbucks fell by 6 percent, but that this change disproportionately affected consumers who made high‐​calorie purchases (thereby potentially having a larger impact on obesity rates).<sup><a href="#cite-17">17</a></sup> Yet a meta‐​analysis by Long et al. found that “current evidence does not support a significant impact on calories ordered.”<sup><a href="#cite-18">18</a></sup> And the findings of Cantor et al. suggest any effects may be short‐​lived. While reduced caloric intake at point‐​of‐​purchase is certainly a necessary condition for reductions in body weight, it is not sufficient. As mentioned previously, the stated goal is to reduce the prevalence of being obese and overweight, especially in the long run. Thus the focus of this study is much more accurately aligned with the explicit public health policy goal of such menu mandates. A recent paper by Deb and Vargas explores many of the same issues as this paper; the authors use the BRFSS and the staggered implementation of menu mandates to examine effects on BMI, although the geographic coverage and econometric methods differ.<sup><a href="#cite-19">19</a></sup> In many respects, the principal findings of the two studies are quite similar: for the population as a whole, the effects of menu mandates on BMI are very small. Deb and Vargas find significant effects for some subgroups, as does this study. And although not the key focus of their study, their entropy‐​balanced, weighted trends for men (where they do find significant effects) show convergence by 2012, consistent with a fade‐​out effect of menu mandates found in this study.</p> <h2><a id="sec-2.2"></a>The Costs of Menu Mandates</h2> <p>Some of the same studies that find reductions in calories consumed (arguably a benefit) also assert that the costs of menu mandates are trivial or nonexistent. For example, Bollinger et al. argue that “as far as regulatory policies go, the costs of calorie posting are very low—so even these small benefits could outweigh the costs.”<sup><a href="#cite-20">20</a></sup> This section reviews the costs of menu mandates.</p> <p>Some of the financial costs are outlined in Bollinger et al. One cost of menu mandates is updating display menus, which is modestly expensive. This potentially is a one‐​time fixed cost, and perhaps a primary reason many chains are switching to digital menu boards.<sup><a href="#cite-21">21</a></sup> Another cost is determining the caloric content of each menu item. This is likely a more important issue for the other types of venues covered by the menu mandate, as most chains know well the caloric content of each regular menu item.<sup><a href="#cite-22">22</a></sup> Related to this, there may be increased legal costs from being exposed to potential litigation if the posted calories are incorrect. Menu mandates may also affect operating profits by decreasing demand or frequency of visits, but this was not the case for Starbucks.<sup><a href="#cite-23">23</a></sup></p> <p>There are also more subtle costs. Adding calorie content can slow down the ordering process, which reduces the overall convenience of consuming fast food. Some menu labeling laws distinguish between—and have different requirements for—menu boards inside a restaurant and drive‐​through menus outside a restaurant. For example, California’s statewide menu mandate (effective January 2011) required menu boards to display calories next to the item, but allowed to drive‐​throughs to offer a brochure that is available on request.<sup><a href="#cite-24">24</a></sup> This law implicitly recognizes the potential bottleneck that arises with one line in a drive‐​through setting, but the same critique about reduced convenience applies inside the store as well.</p> <p>Arguably a more important, but harder to measure, cost is the reduced utility from consuming a meal. Although Cantor et al. find increases of up to 37 percentage points in those who saw calorie labels in New York City after the menu mandate (from 14 percent to 51 percent), those who used labels to order fewer calories increased by just 7 to 10 percentage points. Among those who see such information but do not use it in altering their purchasing choices, such “education” presumably lowers utility for those who still consume high‐​calorie meals anyway. Glaeser calls this an “emotional tax” on behavior that yields no government revenue, just pure utility losses.<sup><a href="#cite-25">25</a></sup></p> <h2><a id="sec-2.3"></a>Empirical Approach and Findings</h2> <p>Although one cannot yet measure the impact of the menu mandate provision in the ACA, a number of localities and states have regulated menu information at chain restaurants since 2008. Most prominently, in New York City under Mayor Michael Bloomberg, efforts were made to regulate soda sizes, limit trans fats, and mandate calorie disclosure on menus, leading to calls of New York becoming a “nanny state.”<sup><a href="#cite-26">26</a></sup> Although receiving far less attention, some of these same measures—especially regarding mandated calorie disclosure—were implemented in a number of other large urban areas, including Philadelphia, Portland, Seattle, as well as statewide in Massachusetts and California. The empirical approach, discussed in the appendix, is to compare individuals in these locations both before and after menu mandates were enforced. To address concerns that other factors besides menu mandates may also affect body weight and were changing over time, other large cities (Charlotte, Chicago, Columbus, Dallas, Denver, Detroit, El Paso, Fort Worth, Houston, Indianapolis, Jacksonville, Louisville, Memphis, Milwaukee, Nashville, Oklahoma City, Phoenix, and San Antonio) serve as a control group.</p> <p>The analysis relies on transparent, publicly available data from the Behavioral Risk Factor Surveillance System. The BRFSS completes more than 400,000 adult interviews each year, making it the largest continuously conducted health survey system in the world.<sup><a href="#cite-27">27</a></sup> Between 2003 and 2012, the publicly available data both identify an individual’s locality and ask about body weight.<sup><a href="#cite-28">28</a></sup> In virtually all studies of adults, the critical outcome of interest is the body mass index, which is a measure of body fat based on height and weight: BMI is a person’s weight in kilograms divided by the square of their height in meters. From there, various thresholds of BMI are used to classify individuals as obese (BMI &gt;= 30.0), overweight (BMI &gt;= 25.0), underweight (BMI &lt; 18.5), or normal weight (18.5 &lt;= BMI &lt; 25.0).<sup><a href="#cite-29">29</a></sup></p> <p>Without a doubt, the largest share of attention has been focused on obesity. More than one‐​third of adults in the United States are obese.<sup><a href="#cite-30">30</a></sup> The empirical analysis in this paper examines the impact of menu mandates on obesity, along with the other body weight outcomes (BMI levels, overweight or more, underweight). Because of the motivations discussed earlier about learning and salience (i.e., the “novelty” of calorie disclosure), this study estimates both the immediate impact and the longer‐​run impact of menu mandates. Figure 1 <a href="#Figure1">Figure 1, “The Effect of Menu Mandates on Obesity Levels is Short‐​Lived”</a> illustrates how menu mandates affect obesity rates for adults in the years after enactment, using coefficient estimates from the regression model based on all 30 cities in Table A.2.<a href="#TableA.2">Table A.2, “Full Sample from the Behavioral Risk Factor Surveillance System”</a></p> <div><a id="Figure1"></a> <p><strong>Figure 1. The Effect of Menu Mandates on Obesity Levels is Short‐​Lived</strong></p> <div> <div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.full" data-entity-type="media" data-entity-uuid="e63e5d18-f996-4a9f-89a1-35edeb293358" data-langcode="en" class="embedded-entity"> <img width="450" height="267" alt="The Effect of Menu Mandates on Obesity Levels is Short-Lived" class="lozad component-image lozad" data-srcset="/sites/cato.org/files/styles/pubs/public/images/figure1_gs_lres.jpg?itok=pNUWPUdJ 1x, /sites/cato.org/files/styles/pubs_2x/public/images/figure1_gs_lres.jpg?itok=QxouFHJW 1.5x" data-src="/sites/cato.org/files/styles/pubs/public/images/figure1_gs_lres.jpg?itok=pNUWPUdJ" typeof="Image" /></div> </div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Source: Effects from regression model were estimated by the author from the Behavioral Risk Factor Surveillance System data in Table 2, Specification 2.</p> </td> </tr></table></div> </div> </div> <p>As can be seen, prior to enactment of menu mandates (period ‑1), approximately 25.7 percent of adults in these 30 major cities were obese. There is a statistically significant reduction in obesity at time of implementation—roughly 1.25 percentage points—which would bring down the obesity rate to 24.5 percent. However, the effects are short‐​lived. In years after enactment, the novelty of menu mandates appears to wear off, and obesity rates again rise, such that the entire impact on obesity disappears within four years. Thus, menu mandates appear to have a small but temporary impact on obesity.</p> <p>In addition to obesity, where the effects fade over time, the study considered BMI, for which there appear to be more permanent effects, although these effects are not concentrated amongst the heaviest individuals. The same empirical models show that menu mandates lead to a one‐​time reduction in BMI of 0.15 BMI points, and that this weight reduction is sustained over time. Figure 2 <a href="#Figure2">Figure 2, “The Impact of Menu Mandates on Body Weight”</a> illustrates the practical importance of this reduction.</p> <div><a id="Figure2"></a> <p><strong>Figure 2. The Impact of Menu Mandates on Body Weight</strong></p> <div> <div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.full" data-entity-type="media" data-entity-uuid="90968734-6039-401b-8bf6-3c7b682c333f" data-langcode="en" class="embedded-entity"> <img width="450" height="296" alt="The Impact of Menu Mandates on Body Weight" class="lozad component-image lozad" data-srcset="/sites/cato.org/files/styles/pubs/public/images/figure2_gs_lres.jpg?itok=JFGY6h8v 1x, /sites/cato.org/files/styles/pubs_2x/public/images/figure2_gs_lres.jpg?itok=sK6NjHfi 1.5x" data-src="/sites/cato.org/files/styles/pubs/public/images/figure2_gs_lres.jpg?itok=JFGY6h8v" typeof="Image" /></div> </div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Source: Author’s calculations from model estimated from the Behavioral Risk Factor Surveillance System data in Table 2, Specification 2. Results were calculated for the average BMI in the sample of 27.3.</p> </td> </tr></table></div> </div> </div> <p>Although the weight loss is statistically significant, an effect of 0.15 BMI points translates into a barely noticeable difference in weight. For example, for an individual who is 5′10″ and is initially average (BMI = 27.3), the reduction in body weight is roughly one pound. As can be seen for heights that vary from 5′0″ to 6′0″, the impact of menu mandates for the typical individual is hardly visible.</p> <p>The technical analysis in the appendix further examines the effects of menu mandates among various socioeconomic groups. It finds that the impacts are nonexistent for young adults and the less educated—both groups where, it could be argued, that such mandates convey new and meaningful information about caloric content. In contrast, the effects (and fade‐​out) are larger for older adults and those with more education—both groups that likely have greater knowledge of caloric content, and where such mandates provide salience, novelty, or guilt when initially implemented. For them, there are larger initial reductions in BMI and obesity, but the initial effects fade out quickly. The conclusion that emerges is that menu mandates serve as an ineffective “emotional tax.”</p> <h2><a id="sec-3"></a>Conclusion</h2> <p>The analysis in this study has found that menu mandates are a futile effort to reduce body weight, with trivial or short‐​lived effects on BMI and obesity. What public efforts should be undertaken to reduce obesity? The intuitive answer is “nothing at all.” People make choices about all aspects of their lives. Whether it is to eat unhealthily, smoke cigarettes, use drugs, consume alcohol, drop out of school, watch too much television, not exercise, or not save for retirement, all of these decisions should ultimately fall onto the individual, who has to live with the consequences of his or her actions. In virtually all of these cases, as illustrated with BMI and obesity in this study, the argument that individuals are ill‐​informed about the consequences of their actions is implausible.</p> <p>Proponents of government intervention would argue that there are negative externalities—costs of obesity that are not borne by the individual, but by society as a whole. The primary consequences of obesity are costs related to disease, medical visits, and lost work days. In principle, each of these would be internalized by the individual through well‐​functioning insurance and labor markets. That is, the fact that Medicare or Medicaid costs increase due to obesity is not a problem about obesity, but about public health insurance not accurately pricing premiums to reflect an individual’s choices. Private, unregulated insurance markets would price their products based on such risk characteristics, in which case such externalities are internalized.</p> <p>Finally, some prominent behavioral economists look at the evidence on ineffectiveness of calorie labeling and suggest doubling down. Cass Sunstein has recently argued that menu mandates are too complicated and argues for “simple and meaningful” disclosures to consumers, such as putting a “red light” on highly caloric foods and a “green light” on the healthier ones.<sup><a href="#cite-31">31</a></sup> The current analysis shows that the problem is not lack of knowledge or conveying information—on the contrary, the consumers who responded to the menu mandates were among the most knowledgeable. Rather, people have preferences that are more or less fixed, and for the most part, people enjoy cheeseburgers more than broccoli. The private market provides ample nutrition advice at extremely low cost, from cell‐​phone apps that give calorie and other nutrition information to easy‐​to‐​understand, simple substitutions in books such as Eat This, Not That. There is no need for government‐​mandated disclosures that impose an emotional tax on each transaction when individuals can easily and voluntarily seek out such information on nutrition.</p> <h2><a id="d5e105"></a>A. Appendix</h2> <h2><a id="sec-4.1"></a>Data</h2> <p>The analysis uses data from the Behavioral Risk Factor Surveillance System.<sup><a href="#cite-32">32</a></sup> The BRFSS is a collaborative project of the Centers for Disease Control and Prevention (CDC) and U.S. states and territories.<sup><a href="#cite-33">33</a></sup> The BRFSS, administered and supported by CDC’s Behavioral Risk Factor Surveillance Branch, is an ongoing data‐​collection program designed to measure behavioral risk factors for the noninstitutionalized adult population (18 years of age and older). The BRFSS was initiated in 1984, with 15 states collecting surveillance data on risk behaviors through monthly telephone interviews. Over time, the number of states participating in the survey increased: by 2001, all 50 states, the District of Columbia, Puerto Rico, Guam, and the U.S. Virgin Islands were participating in the BRFSS.</p> <p>Of critical importance, the BRFSS calculates the body mass index from the respondent’s reported height and weight. The BMI is a measure of body fat based on height and weight that applies to adult men and women, where a BMI of 30.0 or greater is classified as obese, a BMI between 25.0 and 29.9 is classified as overweight, a BMI between 18.5 and 24.9 is classified as normal weight, and a BMI less than 18.5 is classified as underweight.<sup><a href="#cite-34">34</a></sup></p> <p>The BRFSS consists of repeated annual cross sections of randomly sampled adults. The survey boasts a large number of respondents, which is critical to obtaining meaningful precision when examining the impact of a local program where effects might be concentrated amongst only a fraction of the population.<sup><a href="#cite-35">35</a></sup> Given the focus on local regulations regarding caloric content, the analysis uses BRFSS data from 2003 to 2012, where county identifiers are included.<sup><a href="#cite-36">36</a></sup> Adults in the 30 largest cities in the United States are included, reducing the initial BRFSS sample from 3,991,585 observations to 362,361 observations.<sup><a href="#cite-37">37</a></sup> The total population in these cities, approximately 38.97 million in July 2012, is 12.4 percent of the total U.S. population.<sup><a href="#cite-38">38</a></sup> These 30 cities include New York, Philadelphia, Seattle, and Portland, all of which mandated calorie disclosure on menus starting in 2008 or later. The cities also include Los Angeles, San Francisco, San Jose, San Diego, and Boston, where state legislation mandated disclosure. By 2012, nearly half the residents of these 30 large cities were covered by such mandates. The final sample consists of adults aged 18 and over who provided sufficient information to compute BMI; demographics (race, ethnicity, age, gender, number of children, and marital status); socioeconomic status (education, employment, and income); and health status (self‐​reported health and exercise). These restrictions reduce the sample to 288,392 individual‐​level observations that are used in the empirical analysis.</p> <h2><a id="sec-4.2"></a>Summary Statistics</h2> <p>The vast majority of menu mandates were implemented at the local level by very large cities. A natural concern, one that is accounted for in the regression framework with city‐​fixed effects, is that large cities differ from smaller cities or rural areas, and also that large cities with calorie disclosure requirements differ from other ones that did not have such mandates. Table A.1 <a href="#TableA.1">Table A.1, “Comparisons across 30 cities over time in Body Mass Index (BMI), Obesity, and Health Habits”</a> provides summary statistics in 2003 and 2012 across the 30 cities for several key health variables.<sup><a href="#cite-39">39</a></sup> The BMI and obesity (BMI &gt;= 30.0) increased in almost every city over this period. There is significant cross‐​sectional variation in residents’ weight prior to any menu mandates: Detroit, Louisville, and San Antonio had obesity rates exceeding 30 percent in 2003, while many of the cities that subsequently mandated calorie disclosure had obesity rates below 20 percent.</p> <p>Such differences in BMI or obesity could reflect fixed characteristics at the local level, such as the weather (and ease of exercising outdoors) or mode of transport to work, and are controlled for in the empirical work with city‐​fixed effects. Put differently, it is likely that the localities that forced calorie disclosure are different in other ways. This is illustrated in Table A.1 by looking at self‐​reported health and exercise habits, both of which exhibit significant cross‐​sectional variation: in 2003, there is approximately a 20 percentage point difference in reporting any exercise in the past 30 days between the least active and most active cities.</p> <h2><a id="sec-4.3"></a>Empirical Specification</h2> <p>The staggered implementation of menu mandates in some localities, but not others, creates a straightforward “difference‐​in‐​difference” framework that has been effectively used to estimate the causal effect of policy.<sup><a href="#cite-40">40</a></sup> The regression specification is set up as follows:</p> <p>WEIGHTijt = β<sub>0</sub> + β<sub>1</sub>MENU_MANDATE<sub>jt</sub> + β<sub>2</sub>YEARS_AFTER<sub>jt</sub> + β<sub>3</sub>X<sub>i</sub> + δ<sub>j</sub> + δ<sub>t</sub> + ε<sub>ijt</sub></p> <p>where WEIGHT<sub>ijt</sub> represents BMI, Obesity, Overweight, or Underweight for person i in city j (for the 30 cities listed in Table A.1 <a href="#TableA.1">Table A.1, “Comparisons across 30 cities over time in Body Mass Index (BMI), Obesity, and Health Habits”</a>) in time period t (2003–2012), and is a continuous measure for BMI, or a dummy variable equal to 1 if the individual was Obese (BMI &gt;= 30.0), Overweight (BMI &gt;= 25.0) or Underweight (BMI &lt; 18.5). Also included are individual controls in Xi related to the respondent’s age, education, race/​ethnicity, gender, marital status, health status, exercise frequency, and number of children.</p> <p>The variable MENU_MANDATE<sub>jt</sub> is a policy indicator that varies by city and time period, and is equal to 1 if the locality mandated calorie disclosure in year t. Additionally, the variable YEARS_​AFTERjt measures the number of years since the mandate was implemented. At the local level, New York, Philadelphia, Seattle (King County), and Portland (Multnomah County) passed menu mandates.<sup><a href="#cite-41">41</a></sup> At the state level, California, Maine, Massachusetts, and Oregon passed such mandates.<sup><a href="#cite-42">42</a></sup> For example, New York City implemented a mandate in 2008; thus both variables equal 0 for these residents in the years 2003–2007, while MENU_​MANDATEjt equals 1 for the years 2008–2012, and YEARS_​AFTERjt increases from 1 in 2008 to 5 by 2012. The mandate variables are constructed at the group level, while the BRFSS data itself is at the individual level. Following the recommendation of Cameron, Gelbach, and Miller, the standard errors are corrected for non‐​nested two‐​way clustering, where the clustering is based on locality and year.<sup><a href="#cite-43">43</a></sup></p> <div><a id="TableA.1"></a> <p><strong>Table A.1. Comparisons across 30 cities over time in Body Mass Index (BMI), Obesity, and Health Habits</strong></p> <div> <div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.full" data-entity-type="media" data-entity-uuid="c2b5d44c-6841-4331-896e-d76f42d39521" data-langcode="en" class="embedded-entity"> <img width="450" height="581" alt="Media Name: tablea1_gs_lres.jpg" class="lozad component-image lozad" data-srcset="/sites/cato.org/files/styles/pubs/public/images/tablea1_gs_lres.jpg?itok=Vtd8oMlf 1x, /sites/cato.org/files/styles/pubs_2x/public/images/tablea1_gs_lres.jpg?itok=TEiQBACS 1.5x" data-src="/sites/cato.org/files/styles/pubs/public/images/tablea1_gs_lres.jpg?itok=Vtd8oMlf" typeof="Image" /></div> </div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Source: Authors’ tabulation of 2003 and 2012 Behavioral Risk Factor Surveillance System.</p> </td> </tr></table></div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Note: Summary statistics are unweighted and include adults aged 18 and over. Cities with an asterisk had implemented a mandate for chain restaurants to disclose calories by 2012. The BMI is average Body Mass Index and Obese is the fraction with BMI &gt;= 30.0. Good health is the fraction of sample that self‐​reports health status as excellent, very good, or good. Any exercise is the fraction that reports any exercise (outside of work) in the past 30 days. Individuals with invalid data on any question excluded from table.</p> </td> </tr></table></div> </div> </div> <p><strong><a id="sec-4.4"></a>Main Results and Subgroup Analysis</strong></p> <p>Table A.2 presents results for four outcomes: BMI, Obese, Overweight, and Underweight. The first set of results includes an indicator for a menu mandate, but not additional years since passage. Although not shown, all specifications include dummy variables for year and city, interview month, health status, gender, marital status, race/​ethnicity, any exercise, education, and age and number of children. In no case are the results statistically significant. In addition, if one were to interpret the point estimate on BMI, it would indicate that the effect of menu labeling reduces BMI by 0.11 points. The average respondent in the sample has a BMI of 27.3, suggesting such labeling would reduce BMI to approximately 27.2. To put this in perspective, for a 5′10″ male adult, this translates into a reduction in weight from 190 pounds to 189.5 pounds, roughly a 0.5 pound reduction.<sup><a href="#cite-47">47</a></sup> None of the threshold measures—Obese, Overweight, or Underweight—are significant.</p> <div><a id="TableA.2"></a> <p><strong>Table A.2. Full Sample from the Behavioral Risk Factor Surveillance System</strong></p> <div> <div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.full" data-entity-type="media" data-entity-uuid="f5eb9e9d-a4e6-454a-98c3-255b721de615" data-langcode="en" class="embedded-entity"> <img width="450" height="264" alt="Media Name: tablea2_gs_lres.jpg" class="lozad component-image lozad" data-srcset="/sites/cato.org/files/styles/pubs/public/images/tablea2_gs_lres.jpg?itok=VF5CeOyY 1x, /sites/cato.org/files/styles/pubs_2x/public/images/tablea2_gs_lres.jpg?itok=YrdAYijX 1.5x" data-src="/sites/cato.org/files/styles/pubs/public/images/tablea2_gs_lres.jpg?itok=VF5CeOyY" typeof="Image" /></div> </div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Source: Author’s calculation from 2003–2012 Behavioral Risk Factor Surveillance System.</p> </td> </tr></table></div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Notes: Standard errors in parentheses and are corrected for non‐​nested two‐​way clustering, using the methods of A. Colin Cameron, Jonah B. Gelbach, and Douglas L. Miller, “Robust Inference with Multiway Clustering,” Journal of Business &amp; Economic Statistics 29, no. 2 (2011): 238–49, where clustering is grouped on city and year. All specifications includes fixed effects for city (30 overall), year (2003–2012), and interview month. Individual covariates include self‐​reported health (excellent/​very good/good)(omitted is fair/​poor); male; married; race/​ethnicity (Hispanic, white, African-American)(omitted is other group); any exercise in past 30 days; number of children; education (high school or less, some college)(omitted is college graduate); and age.</p> </td> </tr></table></div> </div> </div> <p>The second set of findings examines the full sample, and includes both an indicator for the menu mandate as well as years since passage. For BMI, the initial implementation significantly reduces BMI (p‑value of 0.037). However, the interpretation is much the same as before, as the effect of menu labeling reduces BMI by 0.15 points. The effect appears to be long‐​lived for the full sample, as the effect of years‐​since‐​passage is insignificant.</p> <p>Perhaps the most noteworthy result relates to obesity (BMI &gt;= 30.0). In the full sample, nearly 26 percent of respondents are obese. The immediate impact of menu labeling is to significantly reduce obesity by nearly 1.3 percentage points (p‑value of 0.016). Bollinger et al. argue that if the policy goal is to address obesity, it is important to know whether calorie posting disproportionately affects consumers who make high‐​calorie purchases.<sup><a href="#cite-48">48</a></sup> They find that calorie posting has a large influence on Starbucks loyalty cardholders who tended to make high‐​calorie purchases. For consumers who averaged more than 250 calories per Starbucks transaction, calories per transaction fell by 26 percent, versus 6 percent for the full sample. The short‐​run effect estimated from the BRFSS analysis appears consistent with Bollinger et al.<sup><a href="#cite-49">49</a></sup> However, the effect on obesity is short‐​lived, as the coefficient on years‐​since‐​passage is positive. Each additional year since passage increases obesity by nearly 0.4 percentage points, meaning that the short‐​run reduction in obesity disappears within four years. Menu labeling mandates have no long‐​run impact on obesity. Furthermore, menu mandates have no impact on Overweight (BMI &gt;= 25.0, comprising nearly 61 percent of the full sample) or Underweight (BMI &lt; 18.5, comprising 1.7 percent of the full sample).</p> <p>Table A.3 <a href="#TableA.3">Table A.3, “Learning versus Salience”</a> breaks out the full sample into various subgroups that may be of interest in their own right. Even though the effects of menu mandates are ineffective for the full sample, they may be more significant for various groups. As Bollinger et al. (2011) explain, two of the principal methods through which menu mandates may reduce weight are learning effects and salience effects.<sup><a href="#cite-50">50</a></sup> It is reasonable to believe that the learning effect would be more important for those with less experience with nutrition, where two proxies for such inexperience are low levels of education and young age. Conversely, the learning effect should be less important for those with more experience with nutrition, which is proxied by higher levels of education and older ages. If learning is unimportant for these groups, then any effects on weight are likely due to the salience of caloric information on menus.</p> <div><a id="TableA.3"></a> <p><strong>Table A.3. Learning versus Salience</strong></p> <div> <div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.full" data-entity-type="media" data-entity-uuid="b29e4553-2988-43ac-8c87-d88ac0e4c3dd" data-langcode="en" class="embedded-entity"> <img width="450" height="589" alt="Media Name: tablea3_gs_lres.jpg" class="lozad component-image lozad" data-srcset="/sites/cato.org/files/styles/pubs/public/images/tablea3_gs_lres.jpg?itok=ErYqhV57 1x, /sites/cato.org/files/styles/pubs_2x/public/images/tablea3_gs_lres.jpg?itok=PUNS9iX4 1.5x" data-src="/sites/cato.org/files/styles/pubs/public/images/tablea3_gs_lres.jpg?itok=ErYqhV57" typeof="Image" /></div> </div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Source: Author’s calculation from 2003–2012 Behavioral Risk Factor Surveillance System.</p> </td> </tr></table></div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Notes: Standard errors in parentheses and are corrected for non‐​nested two‐​way clustering, using the methods of A. Colin Cameron, Jonah B. Gelbach, and Douglas L. Miller, “Robust Inference With with Multiway Clustering,” Journal of Business &amp; Economic Statistics29, no. 2 (2011): 238–49, where clustering is grouped on city and year. All specifications includes fixed effects for city (30 overall), year (2003–2012), and interview month. Individual covariates are identical to that in Table 2, except when stratifying on covariate under consideration.</p> </td> </tr></table></div> </div> </div> <p>The first panel of this table examines weight outcomes for those with a high school diploma or less, and the second examines outcomes for young adults aged 18 to 29. For both groups, menu mandates are more likely to convey new information. For less educated individuals, there is no evidence that mandates influence weight, suggesting that the information effect plays a relatively minor role. For young adults, it does appear that mandates reduce obesity, and that such an impact grows over time. When the two groups are combined—less educated young adults—there appears to be some sustained effect on Overweight but no effect on Obesity or Underweight.</p> <p>The second panel examines weight outcomes for respondents with at least some college education, as well as older adults. In these groups, one may speculate that salience of caloric content plays a more important role. The results for more‐​educated respondents are striking. The immediate effect of menu mandates is a BMI reduction of nearly 0.28 BMI points (p‑value of 0.004), but the effect disappears within approximately four years (with BMI increasing by nearly 0.06 BMI points per year, p‑value of 0.053). Menu mandates have long‐​lasting effects on Overweight, but short‐​lasting effects on Obesity. For Overweight, the immediate effect is a reduction of 1.4 percentage points (p‑value of 0.056), and the effect does not diminish over time. For Obesity, there is a large immediate reduction of 1.7 percentage points (p‑value of 0.044), but this effect is eliminated within approximately three years (with Obesity rising by 0.5 percentage points per year, p‑value of 0.042). There is no effect for Underweight. The findings are similar for individuals aged 30 and over. The immediate and long‐​lasting effect on BMI is to reduce it by 0.18 BMI points (p‑value of 0.028). As with more educated individuals, the immediate impact on Obesity is significant but short‐​lived. The immediate reduction is 1.4 percentage points (p‑value of 0.017), but the effect also disappears within three years (with Obesity rising by 0.5 percentage points per year, p‑value of 0.001). As before, effects on Overweight appear to be longer lasting, and there is no effect on Underweight.</p> <p>By combining the two groups—college-educated individuals aged 30 and over—the fade‐​out effects become extremely apparent. The immediate effect of menu mandates reduces BMI by 0.3 BMI points (p‑value of 0.002) but BMI subsequently increases by 0.07 points per year (p‑value of 0.018). There again appear to be sustained effects on Overweight, but effects on Obesity fade out quickly.</p> <p>Table A.4 <a href="#TableA.4">Table A.4, “Intensive Users of Fast Food”</a> breaks outs the sample into those who are likely more intensive users of chain restaurants. Driskell et al. show that a significantly higher percentage of male college students report eating fast foods at least once a week relative to female college students.<sup><a href="#cite-51">51</a></sup> Other work shows that unmarried men spend a significantly greater proportion of their food budget on commercially prepared food than their married male peers. Households headed by single men spent more per capita on such food than those headed by single women.<sup><a href="#cite-52">52</a></sup></p> <p>Given these findings, it is expected that single men would likely be more intensive users of fast food. However, the impact of menu mandates is less clear. It is surely the case that menu mandates should not matter for those who tend to cook at home. Yet among regular users of fast food, it is possible that much of the learning about caloric intake has already been done, or that food choices are relatively ingrained. The findings in the first panel of Table A.4 <a href="#TableA.4" title="Table A.4. Intensive Users of Fast Food">Table A.4, “Intensive Users of Fast Food”</a> show no impact of menu mandates on unmarried men across BMI and each weight category.</p> <div><a id="TableA.4"></a> <p><strong>Table A.4. Intensive Users of Fast Food</strong></p> <div> <div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.full" data-entity-type="media" data-entity-uuid="e0a0d14c-0e34-4a8a-b0f8-dd636191331c" data-langcode="en" class="embedded-entity"> <img width="450" height="145" alt="Media Name: tablea4_gs_lres.jpg" class="lozad component-image lozad" data-srcset="/sites/cato.org/files/styles/pubs/public/images/tablea4_gs_lres.jpg?itok=XSb8QY3H 1x, /sites/cato.org/files/styles/pubs_2x/public/images/tablea4_gs_lres.jpg?itok=htbv68pR 1.5x" data-src="/sites/cato.org/files/styles/pubs/public/images/tablea4_gs_lres.jpg?itok=XSb8QY3H" typeof="Image" /></div> </div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Source: Author’s calculation from 2003–2012 Behavioral Risk Factor Surveillance System.</p> </td> </tr></table></div> <div> <table border="0" summary="Note"><tr><td align="left" valign="top"> <p>Notes: Standard errors in parentheses and are corrected for non‐​nested two‐​way clustering, using the methods of A. Colin Cameron, Jonah B. Gelbach, and Douglas L. Miller, “Robust Inference With with Multiway Clustering,” Journal of Business &amp; Economic Statistics29, no. 2 (2011): 238–49, where clustering is grouped on city and year. All specifications includes fixed effects for city (30 overall), year (2003–2012), and interview month. Individual covariates are identical to that in Table A.2, except when stratifying on covariate under consideration.</p> </td> </tr></table></div> </div> </div> <h2>Notes</h2> <p id="cite-1">1. K. M. Flegal, M. D. Carroll, R. J. Kuczmarski, and C. L. Johnson, “Overweight and Obesity in the United States: Prevalence and Trends, 1960–1994,” <em>International Journal of Obesity and Related Metabolic Disorders: Journal of the International Association for the Study of Obesity</em> 22, no. 1 (1998): 39–47.</p> <p id="cite-2">2. K. M. Flegal, M. D. Carroll, C. L. Ogden, and L. R. Curtin, “Prevalence and Trends in Obesity Among US Adults, 1999–2008.” <em>JAMA</em> 303, no. 3 (2010): 235–41.</p> <p id="cite-3">3. A. H. Mokdad, M. K. Serdula, W. H. Dietz, B. A. Bowman, J. S. Marks, and J. P. Koplan, “The Spread of the Obesity Epidemic in the United States, 1991–1998,” <em>JAMA</em> 282, no. 16 (1999): 1519–22.</p> <p id="cite-4">4. A. M. Wolf and G. A. Colditz, “Current Estimates of the Economic Cost of Obesity in the United States,” <em>Obesity Research</em> 6, no. 2 (1998): 97–106.</p> <p id="cite-5">5. J. Cawley and C. Meyerhoefer, “The Medical Care Costs of Obesity: An Instrumental Variables Approach,” <em>Journal of Health Economics</em> 31, no. 1 (2012): 219–30.</p> <p id="cite-6">6. These include bakeries, cafeterias, coffee shops, convenience stores, delicatessens, food service facilities located within entertainment venues (such as amusement parks, bowling alleys, and movie theatres), food service vendors (e.g., ice cream shops and mall cookie counters), food take‐​out and/​or delivery establishments (such as pizza), grocery stores, retail confectionary stores, superstores, quick service restaurants, and table service restaurants. See Department of Health and Human Services, Food and Drug Administration, “Food Labeling; Nutritional Labeling of Standard Menu Items in Restaurants and Similar Retail Food Establishments,” <em>Federal Register</em> 79, no. 230 (December 1, 2014): 71156, <code class="uri"><a class="uri" href="https://www.gpo.gov/fdsys/pkg/FR-2014-12-01/pdf/2014-27833.pdf" target="_top">https://www.gpo.gov/fdsys/pkg/FR-2014-12-01/pdf/2014-27833.pdf</a></code>.</p> <p id="cite-7">7. See “Health Policy Brief: The FDA’s Menu‐​Labeling Rule,” Health Affairs, June 25, 2015, <code class="uri"><a class="uri" href="http://www.healthaffairs.org/healthpolicybriefs/brief.php?brief_id=140" target="_top">http://www.healthaffairs.org/healthpolicybriefs/brief.php?brief_id=140</a></code>.</p> <p id="cite-8">8. See Stephanie Strom and Sabrina Tavernise, “F.D.A. to Require Calorie Count, Even for Popcorn at the Movies,” <em>New York Times</em>, November 24, 2014, <code class="uri"><a class="uri" href="http://www.nytimes.com/2014/11/25/us/fda-to-announce-sweeping-calorie-rules-for-restaurants.html" target="_top">http://www.nytimes.com/2014/11/25/us/fda-to-announce-sweeping-calorie-rules-for-restaurants.html</a></code>.</p> <p id="cite-9">9. See Department of Health and Human Services, Food and Drug Administration, “Food Labeling; Nutritional Labeling of Standard Menu Items in Restaurants and Similar Retail Food Establishments,” <em>Federal Register</em> 79, no. 230 (December 1, 2014): 71156, <code class="uri"><a class="uri" href="https://www.gpo.gov/fdsys/pkg/FR-2014-12-01/pdf/2014-27833.pdf" target="_top">https://www.gpo.gov/fdsys/pkg/FR-2014-12-01/pdf/2014-27833.pdf</a></code>; and Department of Health and Human Services, Food and Drug Administration, “Food Labeling; Nutritional Labeling of Standard Menu Items in Restaurants and Similar Retail Food Establishments; Extension of Compliance Date,” RIN 0910-AG57, (December 1, 2015), <code class="uri"><a class="uri" href="https://s3.amazonaws.com/public-inspection.federalregister.gov/2015-16865.pdf" target="_top">https://s3.amazonaws.com/public-inspection.federalregister.gov/2015-16865.pdf</a></code>.</p> <p id="cite-10">10. Food and Drug Administration, “Food Labeling; Nutrition Labeling of Standard Menu Items in Restaurants and Similar Retail Food Establishments; Calorie Labeling of Articles of Food in Vending Machines; Final Rule,” <em>Federal Register</em> 79, no. 230 (2014): 71156–259.</p> <p id="cite-11">11. See Jane Furse, “Mayor Bloomberg’s 2008 Calorie‐​posting Has Caused Drop in High‐​calorie Intake: Reports,” <em>New York Daily News</em>, February 28, 2011, <code class="uri"><a class="uri" href="http://www.nydailynews.com/new-york/mayor-bloomberg-2008-calorie-posting-caused-drop-high-calorie-intake-reports-article-1.139154/" target="_top"> http://www.nydailynews.com/new-york/mayor-bloomberg-2008-calorie-posting-caused-drop-high-calorie-intake-reports-article-1.139154</a></code>.</p> <p id="cite-12">12. B. Bollinger, P. Leslie, and A. Sorensen, “Calorie Posting in Chain Restaurants,” <em>American Economic Journal: Economic Policy</em> 3, no. 1 (2011): 91–128.</p> <p id="cite-13">13. Ibid.</p> <p id="cite-14">14. See, for example, R. J. Reynolds Tobacco Company, et al. v. Food and Drug Administration, et al., 11 F.3d 5332 (D.C. Cir. 2012), <code class="uri"><a class="uri" href="https://www.cadc.uscourts.gov/internet/opinions.nsf/4C0311C78EB11C5785257A64004EBFB5/$file/11-5332-1391191.pdf" target="_top">https://www.cadc.uscourts.gov/internet/opinions.nsf/4C0311C78EB11C5785257A64004EBFB5/$file/11-5332-1391191.pdf</a></code>, where the Court concluded “FDA has not provided a shred of evidence … showing that the graphic warnings will ‘directly advance’ its interest in reducing the number of Americans who smoke.” See J. P. Benway, “Banner Blindness: The Irony of Attention Grabbing on the World Wide Web,” <em>Proceedings of the Human Factors and Ergonomics Society Annual Meeting</em> 42, no. 5 (1998): 463–67.</p> <p id="cite-15">15. Jonathan Cantor, Alejandro Torres, Courtney Abrams, and Brian Elbel, “Five Years Later: Awareness of New York City’s Calorie Labels Declined, With No Changes in Calories Purchased,” <em>Health Affairs</em> 34, no. 11 (2015): 1893–900.</p> <p id="cite-16">16. Bollinger et al., “Calorie Posting in Chain Restaurants.”</p> <p id="cite-17">17. Other studies often rely on empirical methods that potentially suffer from selection bias or omitted variables bias. For example, Mary Basset et al. examine purchasing behavior of restaurant patrons, and find that those who buy food at Subway (which posted calorie information at point of purchase) and viewed the calorie information purchased 52 fewer calories than did other Subway patrons (713.8 calories versus 765.5 calories). Of course, those with healthier eating habits may be both more likely to seek calorie information and make healthier purchases. See Mary T. Bassett, Tamara Dumanovsky, Christina Huang, Lynn D. Silver, Candace Young, Cathy Nonas, Thomas D. Matte, Sekai Chideya, and Thomas R. Frieden, “Purchasing Behavior and Calorie Information at Fast‐​Food Chains in New York City, 2007,” American Journal of Public Health 98, no. 8 (August 2008): 1457–59, <code class="uri"><a class="uri" href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2446463/" target="_top">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2446463/</a></code>.</p> <p id="cite-18">18. Matthew W. Long, Deirdre K. Tobias, Angie L. Cradock, Holly Batchelder, and Steven L. Gortmaker, “Systematic Review and Meta‐​analysis of the Impact of Restaurant Menu Calorie Labeling,” <em>American Journal of Public Health</em> 105, no. 5 (2015): E11-24.</p> <p id="cite-19">19. P. Deb and C. Vargas, “Who Benefits from Calorie Labeling? An Analysis of Its Effect on Body Mass,” NBER Working Paper no. 21992 (February 2016).</p> <p id="cite-20">20. Bollinger et al., “Calorie Posting in Chain Restaurants.”</p> <p id="cite-21">21. See Digital Signage Federation, “Primary Benefits of Digital Signage/​Menu Boards for Restaurants and QSR Locations,” <code class="uri"><a class="uri" href="http://www.digitalsignagefederation.org/Resources/Documents/Articles%20and%20Whitepapers/Primary_Benefits_DSMenuBoards.pdf" target="_top">http://www.digitalsignagefederation.org/Resources/Documents/Articles%20and%20Whitepapers/Primary_Benefits_DSMenuBoards.pdf</a></code>, where the first benefit listed is “Ease of Menu Changes to Accommodate New Local, State, Federal &amp; International Menu Board Regulations Regarding Ingredients.”.</p> <p id="cite-22">22. See U.S. Food and Drug Administration, “Questions and Answers on the Menu and Vending Machines Nutrition Labeling Requirements,” December 2, 2015, <code class="uri"><a class="uri" href="http://www.fda.gov/Food/IngredientsPackagingLabeling/LabelingNutrition/ucm248731.htm" target="_top">http://www.fda.gov/Food/IngredientsPackagingLabeling/LabelingNutrition/ucm248731.htm</a></code> for examples of items that would have to be covered.</p> <p id="cite-23">23. Bollinger et al., “Calorie Posting in Chain Restaurants.”</p> <p id="cite-24">24. See “An Act to Add Section 114094 to the Health and Safety Code, Relating to Food Facilities,” California SB 1420 (2008), <code class="uri"><a class="uri" href="http://www.leginfo.ca.gov/pub/07-08/bill/sen/sb_14011450/sb_1420_bill_20080903_enrolled.html" target="_top">http://www.leginfo.ca.gov/pub/07-08/bill/sen/sb_1401-1450/sb_1420_bill_20080903_enrolled.html</a></code>.</p> <p id="cite-25">25. E. L. Glaeser, “Paternalism and Psychology,” <em>University of Chicago Law Review</em> 73, no. 1 (2006): 133–56.</p> <p id="cite-26">26. See “Mayor Bloomberg, Deputy Mayor Gibbs and Health Commissioner Farley Release New Data Highlighting Strong Relationship between Sugary Drink Consumption and Obesity,” City of New York, March 11, 2013, <code class="uri"><a class="uri" href="http://www1.nyc.gov/office-of-the-mayor/news/088-13/mayor-bloomberg-deputy-mayor-gibbs-health-commissioner-farley-release-new-data-highlighting" target="_top">http://www1.nyc.gov/office-of-the-mayor/news/088-13/mayor-bloomberg-deputy-mayor-gibbs-health-commissioner-farley-release-new-data-highlighting</a></code>; Michael Bloomberg, “Mayor Bloomberg Signs Legislation Reinforcing Board of Health’s Trans Fat Restriction,” remarks, New York City, March 28, 2007, <code class="uri"><a class="uri" href="http://www.nyc.gov/portal/site/nycgov/menuitem.c0935b9a57bb4ef3daf2f1c701c789a0/index.jsp?pageID=mayor_press_release&amp;catID=1194&amp;doc_name=http%3A%2F%2Fwww.nyc.gov%2Fhtml%2Fom%2Fhtml%2F2007a%2Fpr091-07.html&amp;cc=unused1978&amp;rc=1194&amp;ndi=1" target="_top">http://www.nyc.gov/portal/site/nycgov/menuitem.c0935b9a57bb4ef3daf2f1c701c789a0/index.jsp?pageID=mayor_press_release&amp;catID=1194&amp;doc_name=http%3A%2F%2Fwww.nyc.gov%2Fhtml%2Fom%2Fhtml%2F2007a%2Fpr091-07.html&amp;cc=unused1978&amp;rc=1194&amp;ndi=1</a></code>; “Mayor Bloomberg, Public Advocate Deblasio, Manhattan Borough President Stringer, Montefiore Hospital CEO Safyer, Deputy Mayor Gibbs, and Health Commissioner Farley Highlight Health Impacts of Obesity,” City of New York, June 5, 2012, <code class="uri"><a class="uri" href="http://www.nyc.gov/portal/site/nycgov/menuitem.c0935b9a57bb4ef3daf2f1c701c789a0/index.jsp?pageID=mayor_press_release&amp;catID=1194&amp;doc_name=http%3A%2F%2Fwww.nyc.gov%2Fhtml%2Fom%2Fhtml%2F2012a%2Fpr200-12.html&amp;cc=unused1978&amp;rc=1194&amp;ndi=1" target="_top">http://www.nyc.gov/portal/site/nycgov/menuitem.c0935b9a57bb4ef3daf2f1c701c789a0/index.jsp?pageID=mayor_press_release&amp;catID=1194&amp;doc_name=http%3A%2F%2Fwww.nyc.gov%2Fhtml%2Fom%2Fhtml%2F2012a%2Fpr200-12.html&amp;cc=unused1978&amp;rc=1194&amp;ndi=1</a></code>; Karen Harned, “The Michael Bloomberg Nanny State in New York: A Cautionary Tale,” <em>Forbes</em>, May 10, 2013, <code class="uri"><a class="uri" href="http://www.forbes.com/sites/realspin/2013/05/10/the-michael-bloomberg-nanny-state-in-new-york-a-cautionary-tale/%23d2424681c6a5/" target="_top">http://www.forbes.com/sites/realspin/2013/05/10/the-michael-bloomberg-nanny-state-in-new-york-a-cautionary-tale/#d2424681c6a5/</a></code>.</p> <p id="cite-27">27. See Centers for Disease Control and Prevention, “Behavioral Risk Factor Surveillance System,” February 1, 2016, <code class="uri"><a class="uri" href="http://www.cdc.gov/brfss/" target="_top">http://www.cdc.gov/brfss/</a></code>.</p> <p id="cite-28">28. The CDC reports that one data limitation of the Behavioral Risk Factor Surveillance System is that “Reliance on self‐​reported heights and weights to calculate the BMI is likely to underestimate average BMI and the proportion of the population in higher BMI categories in population surveys.” See Centers for Disease Control and Prevention, “Diabetes Public Health Resource: Methods and Limitations,” October 15, 2014, <code class="uri"><a class="uri" href="http://www.cdc.gov/diabetes/statistics/comp/methods.htm" target="_top">http://www.cdc.gov/diabetes/statistics/comp/methods.htm</a></code>. It is possible that the small, temporary effects of menu mandates are due to temporarily changing social norms or awareness, rather than an actual temporary reduction in body weight. There is little reason to think, however, that people that people in cities with menu mandates systematically differ in their reporting of weight compared to those in other cities. Also, even if people understate their weight in levels, it isn’t clear that changes in weight (i.e., from the difference‐​in‐​difference framework) will be affected. If everyone reports their weight as 5 lbs. lower than it really is, the change in body weight pre/​post menu mandate will be unaffected. In either case, the effects fade out quickly.</p> <p id="cite-29">29. See Centers for Disease Control and Prevention, Division of Nutrition, Physical Activity, and Obesity, “About Adult BMI,” May 15, 2015, <code class="uri"><a class="uri" href="http://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/" target="_top">http://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/</a></code>.</p> <p id="cite-30">30. See Centers for Disease Control and Prevention, Division of Nutrition, Physical Activity, and Obesity, “Adult Obesity Facts,” September 21, 2015, <code class="uri"><a class="uri" href="http://www.cdc.gov/obesity/data/adult.html" target="_top">http://www.cdc.gov/obesity/data/adult.html</a></code>.</p> <p id="cite-31">31. See Cass Sunstein, “Don’t Give Up on Fast‐​Food Calorie Labels,” <em>Bloomberg</em> View, November 3, 2013, <code class="uri"><a class="uri" href="http://www.bloombergview.com/articles/2015-11-03/don-t-give-up-on-fast-food-calorie-labels" target="_top">http://www.bloombergview.com/articles/2015-11-03/don-t-give-up-on-fast-food-calorie-labels</a></code>.</p> <p id="cite-32">32. The Annual Survey Data is publicly available at Centers for Disease Control and Prevention, “The Behavioral Risk Factor Surveillance System,” September 15, 2015, <code class="uri"><a class="uri" href="http://www.cdc.gov/brfss/annual_data/annual_data.htm" target="_top">http://www.cdc.gov/brfss/annual_data/annual_data.htm</a></code>.</p> <p id="cite-33">33. This description comes directly from Centers for Disease Control and Prevention, “Behavioral Risk Factor Surveillance System Overview 2012,” July 15, 2013, <code class="uri"><a class="uri" href="http://www.cdc.gov/brfss/annual_data/2012/pdf/overview_2012.pdf" target="_top">http://www.cdc.gov/brfss/annual_data/2012/pdf/overview_2012.pdf</a></code>.</p> <p id="cite-34">34. See U.S. Department of Health and Human Services, National Institutes of Health, “Calculate Your Body Mass Index,” <code class="uri"><a class="uri" href="http://www.nhlbi.nih.gov/health/educational/lose_wt/BMI/bmicalc.htm" target="_top">http://www.nhlbi.nih.gov/health/educational/lose_wt/BMI/bmicalc.htm</a></code>.</p> <p id="cite-35">35. C. Courtemanche and D. Zapata, “Does Universal Coverage Improve Health? The Massachusetts Experience,” <em>Journal of Policy Analysis and Management</em> 33, no. 1 (2014): 39–69.</p> <p id="cite-36">36. The 2013 and 2014 BRFSS data does not include county identifiers.</p> <p id="cite-37">37. See U.S. Census Bureau, “American Fact Finder,” <code class="uri"><a class="uri" href="http://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml" target="_top">http://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml</a></code>. The cities include New York; Los Angeles; Chicago; Houston; Philadelphia; Phoenix; San Antonio; San Diego; Dallas; San Jose; Austin; Jacksonville; Indianapolis; San Francisco; Columbus; Fort Worth; Charlotte; Detroit; El Paso; Memphis; Boston; Seattle; Denver; Washington, D.C.; Nashville; Baltimore; Louisville; Portland; Oklahoma City; and Milwaukee.</p> <p id="cite-38">38. U.S. Census Bureau, Population Division, “Annual Estimates of the Resident Population for Incorporated Places Over 50,000, Ranked by July 1, 2012, Population: April 1, 2010 to July 1, 2012,” May 2013.</p> <p id="cite-39">39. All statistics are unweighted. Other characteristics, such as age, vary within the sample over time. Such characteristics are controlled for in the regressions.</p> <p id="cite-40">40. See, for example, Aaron Yelowitz, “The Medicaid Notch, Labor Supply, and Welfare Participation: Evidence from Eligibility Expansions.” <em>Quarterly Journal of Economics</em> 110, no. 4 (1995): 909–39; James Marton, Aaron Yelowitz, and Jeffrey C. Talbert, “A Tale of Two Cities? The Heterogeneous Impact of Medicaid Managed Care,” <em>Journal of Health Economics</em> 36, no. 1 (July 2014): 47–68; and James Marton and Aaron Yelowitz, “Health Insurance Generosity and Conditional Coverage: Evidence from Medicaid Managed Care in Kentucky,” <em>Southern Economic Journal</em> 82, no. 2 (October 2015): 535–55.</p> <p id="cite-41">41. See “Menu‐​Labeling Laws,” Menu‐​Calc, <code class="uri"><a class="uri" href="http://www.menucalc.com/menulabeling.aspx" target="_top">http://www.menucalc.com/menulabeling.aspx</a></code>. Nashville passed, but did not implement, a menu mandate. Several other counties in New York passed menu mandates, but none of the cities in those counties are in the top 30 cities.</p> <p id="cite-42">42. With the exception of Portland, Oregon, neither Maine’s law nor Oregon’s law affected any of the 30 largest cities.</p> <p id="cite-43">43. A. Colin Cameron, Jonah B. Gelbach, and Douglas L. Miller, “Robust Inference with Multiway Clustering,” <em>Journal of Business &amp; Economic Statistics</em> 29, no. 2 (2011): 238–49.</p> <p id="cite-44">44. Bollinger et al., “Calorie Posting in Chain Restaurants.”</p> <p id="cite-45">45. Ibid.</p> <p id="cite-46">46. In addition, all specifications have been estimated also including city‐​specific time trends. The key substantive conclusion—that such mandates have small or insignificant effects on BMI and other outcomes—holds even more forcefully with the inclusion of such trends.</p> <p id="cite-47">47. See U.S. Department of Health and Human Services, National Institutes of Health, “Calculate Your Body Mass Index,” <code class="uri"><a class="uri" href="http://www.nhlbi.nih.gov/health/educational/lose_wt/BMI/bmicalc.htm" target="_top">http://www.nhlbi.nih.gov/health/educational/lose_wt/BMI/bmicalc.htm</a></code>.</p> <p id="cite-48">48. Bollinger et al., “Calorie Posting in Chain Restaurants.”</p> <p id="cite-49">49. Ibid.</p> <p id="cite-50">50. Ibid.</p> <p id="cite-51">51. J. A. Driskell, B. R. Meckna, and N. E. Scales, “Differences Exist in the Eating Habits of University Men and Women at Fast‐​food Restaurants,” <em>Nutrition Research</em> 26, no. 10 (2006): 524–30.</p> <p id="cite-52">52. E. Kroshus, “Gender, Marital Status, and Commercially Prepared Food Expenditure,” <em>Journal of Nutrition Education and Behavior</em> 40, no. 6 (2008): 355–60.</p> </div> Wed, 13 Apr 2016 10:26:00 -0400 Aaron Yelowitz https://www.cato.org/publications/policy-analysis/menu-mandates-obesity-futile-effort Obamacare’s Not‐​So‐​Hidden Tax: Thank You for Smoking https://www.cato.org/blog/obamacares-not-so-hidden-tax-thank-you-smoking Aaron Yelowitz <p>Without government interference, insurance markets will naturally charge higher premiums for riskier individuals. For example, <a href="https://www.cato.org/life%20insurance">life insurance</a> premiums vary considerably based on factors that increase the likelihood of death, such as age, gender, smoking status, and health.&#13;<br /> &#13;<br /> Under Obamacare, many factors that influence healthcare expenditures are excluded from premiums. For example, premiums make no distinction for obesity, likelihood of having a baby, alcoholism or pre-existing conditions. One notable exception is for smokers, where premiums may be up to 50 percent higher than that for non-smokers. I have collected data on premiums for smokers and non-smokers in 35 states, and the data shows large variation in the extent to which smokers are charged more for their choice.&#13;<br /> &#13;</p> <p>Smokers are certainly a risker group than non-smokers. Thus, one would expect some actuarial adjustment to premiums. Given the variation across states, it is clear that premiums vary not only due to a smoker’s greater risk, but other factors as well. At least part of the markup for smokers should be viewed as a “smoker’s tax” rather than an actuarial adjustment.&#13;<br /> &#13;<br /> One expects that the detrimental effects of smoking would build over time. You wouldn’t expect to see large risk adjustments for young individuals. Let’s consider a 27-year-old who doesn’t receive subsidies but is mandated to purchase health insurance. If a non-smoker lived in Cheyenne, WY, he or she could purchase <a href="http://go.hc.gov/1LqgZDv">Blue Cross Blue Shield of Wyoming - BlueSelect Silver ValueTwo Plus Dental</a> plan for $334 per month. This plan has a $3,000 deductible and an out-of-pocket maximum of $6,600. If the 27-year-old smoked, the <a href="http://go.hc.gov/1ObD348">same plan</a> would be $417 per month, or 24.9% higher. For a pack-a-day smoker, this represents a $2.72 per-pack increase in expenditure due to Obamacare.&#13;<br /> &#13;</p> <table><tbody><tr><td>&#13;<br /> &#13; <p><strong>Smoker’s Premium in Wyoming - $2.72 per pack for a pack-a-day 27-year-old smoker</strong></p> <p>&#13; </p> </td> </tr><tr><td>Smoker</td> </tr><tr><td> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="536c4dd5-1a8f-45e4-b7ba-688bb854984a" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/smoker_-_wy.jpg?itok=dWC9LhlF 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/smoker_-_wy.jpg?itok=uS-EbIzg 1.5x" width="700" height="376" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/smoker_-_wy.jpg?itok=dWC9LhlF" alt="Media Name: smoker_-_wy.jpg" typeof="Image" class="component-image" /></p></div> </td> </tr><tr><td>Non-Smoker</td> </tr><tr><td> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="6b810cd8-1a90-4b17-9045-cff05fa4513b" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/non-smoker_-_wy.jpg?itok=YlDU2i0L 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/non-smoker_-_wy.jpg?itok=q_4bC8Id 1.5x" width="700" height="375" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/non-smoker_-_wy.jpg?itok=YlDU2i0L" alt="Media Name: non-smoker_-_wy.jpg" typeof="Image" class="component-image" /></p></div> </td> </tr></tbody></table><p> &#13;<br /> &#13;<br /> However, not all of the $2.72 per-pack is a tax. Smokers are more expensive. Consider a non-smoker in Marquette, MI, who selects <a href="http://go.hc.gov/1ObHX0Y">Blue Cross Blue Shield of Michigan - Blue Cross Silver Extra with Dental and Vision, a Multi-State Plan</a>. That person pays $335 per month, nearly identical to the premium for the non-smoker in Wyoming. The plan has a $2,000 deductible and an out-of-pocket maximum of $5,500. If the 27-year smoked, the <a href="http://go.hc.gov/1Lqqfrs">same plan</a> costs $351 per month (4.8 percent higher), or $0.53 per-pack of cigarettes. If 53 cents per pack approximates the actuarial adjustment for young smoker, then much of the mark-up in Wyoming – $2.19 of the $2.72 – doesn’t represent risk, and can be viewed as a smoking tax.&#13;<br /> &#13;</p> <table><tbody><tr><td>&#13;<br /> &#13; <p><strong>Smoker’s Premium in Michigan - $0.53 per pack for a pack-a-day 27-year-old smoker</strong></p> <p>&#13; </p> </td> </tr><tr><td>Smoker</td> </tr><tr><td> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="76a17283-30d0-4266-bcd7-df98e5d47d03" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/smoker_-_mi.jpg?itok=dv6PBS4N 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/smoker_-_mi.jpg?itok=xjABvQiM 1.5x" width="700" height="376" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/smoker_-_mi.jpg?itok=dv6PBS4N" alt="Media Name: smoker_-_mi.jpg" typeof="Image" class="component-image" /></p></div> </td> </tr><tr><td>Non-Smoker</td> </tr><tr><td> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="b750ada4-7bbc-4186-9838-f02923bf8dc8" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/non-smoker_-_mi.jpg?itok=5KbTuJOT 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/non-smoker_-_mi.jpg?itok=qUm2Yiym 1.5x" width="700" height="375" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/non-smoker_-_mi.jpg?itok=5KbTuJOT" alt="Media Name: non-smoker_-_mi.jpg" typeof="Image" class="component-image" /></p></div> </td> </tr></tbody></table><p> &#13;<br /> &#13;<br /> It might be the case that the numbers above are the exception, not the rule. Yet, in a more comprehensive analysis of premiums, it is clear that the smoker’s premium varies considerably by state. Wyoming has some of the highest mark-ups, while Michigan has some of the lowest mark-ups. The plans presented above are quite similar with respect to premiums and cost sharing for non-smokers, yet the smoker’s mark-up varies greatly. The table below shows average mark-ups for young smokers, restricting the set of plans to Obamacare “Silver” plans for 27-year-olds.&#13;<br /> &#13;<br />  </p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="deb9764a-78dd-4703-90f5-ddc3c8ae5da8" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/smoking_markups.jpg?itok=RmmYYvGs 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/smoking_markups.jpg?itok=PkaynH8u 1.5x" width="700" height="355" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/smoking_markups.jpg?itok=RmmYYvGs" alt="Media Name: smoking_markups.jpg" typeof="Image" class="component-image" /></p></div> <p>If a “pack-a-day” smoker is <a href="http://www.cnn.com/2011/HEALTH/03/15/pack.smokers.now.rare/">an overstatement</a> for actual consumption, then Obamacare cigarette taxes are extremely high in some states – in fact, far higher than the explicit excise tax. The median excise tax on cigarettes is <a href="http://www.taxadmin.org/fta/rate/cigarette.pdf">$1.36</a> per pack, and is $0.60 per pack in Wyoming (ranking 40<sup>th</sup> out of the states) and $2.00 per pack in Michigan (ranking 12<sup>th</sup>).&#13;<br /> &#13;<br /> Based on this analysis, it is clear that the Obamacare smoker’s tax doesn’t represent risk adjustment in many states. But why are cigarette taxes in Obamacare – above and beyond the actuarial adjustment – a problem? Aren’t smokers are doing something terrible to themselves (and others, through secondhand smoke)? In economics, one of the core assumptions is individual rationality. People weight the costs and benefits of their actions and do what’s best for them. Everyday behavior – from smoking cigarettes, to eating pizza instead of broccoli (<a href="https://www.youtube.com/watch?v=8W6rntBADUQ">or sometimes both</a>), to jaywalking in order to save a few seconds of time, to getting in the car to drive to work – involves risk and rewards. If people understand the inherent risks and rewards, then we respect consumer autonomy even if we wouldn’t make the same choice. The economic argument for taxing behavior like smoking (through excise taxes or Obamacare taxes) is that it creates negative externalities. For smoking, there are in fact <a href="http://eml.berkeley.edu/~saez/course131/externalities2_ch06.pdf">negative externalities</a>. These are costs produced but not borne by the smoker, the most obvious of which is secondhand smoke. When such externalities exist, corrective taxation is one of several ways that a more efficient allocation of resources can be achieved. Nonetheless, evidence suggests that cigarette taxes at their current levels more than pay for such negative externalities. As importantly, there’s no reason to think these externalities are much different in Wyoming and Michigan.&#13;<br /> &#13;<br /> With that said, should we be concerned with Obamacare cigarette taxes versus, say, excise taxes? One disadvantage of differing excise taxes across state or city borders is that it encourages smuggling or purchases from low-tax areas. Thus, the tax doesn’t correct the negative externality. That differs, of course, from Obamacare taxes where a person would need to move from Wyoming to Michigan to reduce the tax. Yet, Obamacare cigarette taxes present a host of problems. The vast majority of people do not receive health insurance from Obamacare, so its cigarette taxes do not correct the externalities smoking produces. In addition, the cigarette taxes in Obamacare lack transparency. They are buried in the weeds of Obamacare premiums as hefty smoking taxes, meant to influence or punish the choices of <a href="http://www.cdc.gov/tobacco/data_statistics/fact_sheets/adult_data/cig_smoking/">18 percent</a> of American adults. Perhaps if smoking rates were as high as <a href="http://www.cdc.gov/nchs/fastats/obesity-overweight.htm">obesity</a>, they’d have enough political power that bureaucrats wouldn’t punish them.&#13;<br /> &#13;<br />  &#13;<br /> &#13;<br /><a href="http://www.Yelowitz.com"><em>Aaron Yelowitz</em></a><em> is an associate professor in economics at University of Kentucky and a Visiting Scholar at Cato Institute.</em></p> Fri, 17 Jul 2015 15:19:00 -0400 Aaron Yelowitz https://www.cato.org/blog/obamacares-not-so-hidden-tax-thank-you-smoking The Obamacare Giveaway, Connecticut Edition: Earn $62k and Get Health Insurance for less than $58/​Year https://www.cato.org/blog/obamacare-giveaway-connecticut-edition-earn-62k-get-health-insurance-less-58year Aaron Yelowitz <p>Several days ago, I pointed out that a married couple earning $62,000 in <a href="https://www.cato.org/blog/obamacare-giveaway-wisconsin-edition-earn-62k-get-free-insurance">Wisconsin</a> could get health insurance under Obamacare with no monthly premiums. Now it’s time to move onto Connecticut. Connecticut runs its own exchange, known as <a href="https://www.accesshealthct.com/AHCT/LandingPageCTHIX">access health CT</a>. <br /><br /><br /> Among the 114,000 individuals aged 55 to 64 in Fairfield County, Connecticut, roughly two‐​thirds – or 77,000 people – rely on employer coverage, where the odds are high that they’re paying something out of their own pocket for monthly premiums. <br /><br /><br /> Consider married couple earning $62,000. Each is 64‐​years‐​old, a non‐​smoker, and lives in Fairfield County, Connecticut. The structure of Obamacare subsidies means that many individuals who are not poor can find health plans with such large subsidies that they pay virtually nothing for premiums out of their own pocket. In this case, the couple would qualify for the HealthyCT Bronze Basic HSA 1 Plan for $4.79 per month in premium – or $2.40 per month for each person in that household. If the couple chooses this plan, it pays less than 0.1% of its total income towards health care premiums. That’s not a typo – and it doesn’t say one percent – the household would pay one‐​tenth of one percent of their income towards premiums. Subsidies pay for more than 99% of the monthly premium. <br /><br /><br /> See the graphic below for this married couple:</p> <table><tbody><tr><td>&#13;<br /><p>Married Couple, Both Age 64, Earning $62,000, Fairfield County, CT</p> <p>Pay $57.48 per year for premiums</p> </td> </tr></tbody></table><div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="cd4588ad-0dc6-4003-9f6e-715ed8c9b9c0" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/fairfield_county_ct_age64_age64_62000.jpg?itok=bQhcGOcY 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/fairfield_county_ct_age64_age64_62000.jpg?itok=JWqVCucl 1.5x" width="700" height="376" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/fairfield_county_ct_age64_age64_62000.jpg?itok=bQhcGOcY" alt="Media Name: fairfield_county_ct_age64_age64_62000.jpg" typeof="Image" class="component-image" /></p></div> <p>Although plenty of other plans exist – with higher premiums but less cost sharing – this plan is essentially a giveaway for the near‐​elderly who didn’t want to purchase coverage, but were mandated to do so by the government. <br /><br /><br /> It is also the case that it is sometimes <a href="https://www.cato.org/blog/obamacare-giveaway-its-better-be-64-30-sometimes">better to be 64 than 30</a> in Connecticut. For a couple where both are age 30 (instead of 64), but otherwise identical (earning $62,000, living in the same location, and choose the identical plan), they’d pay $3,441 per year for the same plan or 5.5% of their total income. See the graphic below for this younger couple: <br /></p> <table><tbody><tr><td>&#13;<br /><p>Married Couple, Both Age 30, Earning $62,000, Fairfield County, CT</p> <p>Pay $3,441 per year for premiums</p> </td> </tr></tbody></table><div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="3b487e3f-6f64-4d56-97fc-c25dc9425ace" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/fairfield_county_ct_age30_age30_62000.jpg?itok=d88qoh2t 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/fairfield_county_ct_age30_age30_62000.jpg?itok=tFUsz8BQ 1.5x" width="700" height="376" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/fairfield_county_ct_age30_age30_62000.jpg?itok=d88qoh2t" alt="Media Name: fairfield_county_ct_age30_age30_62000.jpg" typeof="Image" class="component-image" /></p></div> <p>These sorts of findings – both with respect to the relatively high income thresholds where a couple (or individual) gets near‐​free coverage and where younger people in identical situations pay more for the exact same plan – are a prominent feature in Obamacare, and relates to the unusual structure of the subsidy. As mentioned <a href="https://www.cato.org/blog/obamacare-giveaway-its-better-be-64-30-sometimes">previously</a>, these findings related to the fact that the subsidy is pegged to the more generous second‐​lowest‐​cost silver plan rather than the most efficient plan in terms of premiums. <br /><br /><br /> The Connecticut example presents two public policy lessons. First, it is important to realize just how large the redistribution can be. The near‐​elderly married couple near 400% of the poverty line pays virtually nothing out of pocket for health insurance premiums in this example. <a href="http://aspe.hhs.gov/health/reports/2014/premiums/2014mktplaceprembrf.pdf">Government reports</a> emphasize affordability (such as noting that 46% of individuals pay less than $600 per year for premiums), but they do not emphasize nearly as much the near‐​free, giveaway nature that many people face. <br /><br /><br /> Second, it should be noted that the free plans are almost always high deductible plans. If consumers are more careful shoppers of healthcare services when they initially face the actual out‐​of‐​pocket cost of those services (such as with these bronze plans until they reach the deductible), this will enhance efficiency. Offsetting this, however, is the fact that some healthcare services can be postponed, and the consumer can migrate later on to more generous plans when they anticipate greater use of the healthcare system. This is discussed in <a href="http://yelowitz.com/Yelowitz_fertility.pdf">Apostolova and Yelowitz (2015)</a> with respect to Massachusetts health reform in the 2006; they note that women can move in and out of generous health plans easily, which would allow them to purchase more comprehensive coverage when anticipating pregnancy. To the extent that individuals enroll in high deductible plans until they get sick (and then migrate to more generous plans), the efficiency gains under the current Obamacare design are likely to be small. <br /><br /><br /><a href="http://www.yelowitz.com/"><em>Aaron Yelowitz</em></a><em> is an associate professor in economics at University of Kentucky and a Visiting Scholar at Cato Institute.</em>&#13;<br /><br /><br /> </p> Thu, 09 Jul 2015 12:30:53 -0400 Aaron Yelowitz https://www.cato.org/blog/obamacare-giveaway-connecticut-edition-earn-62k-get-health-insurance-less-58year The Obamacare Giveaway – It’s Better to Be 64 than 30 (sometimes) https://www.cato.org/blog/obamacare-giveaway-its-better-be-64-30-sometimes Aaron Yelowitz <p>Take a typical 30-year-old and 64-year-old, earning identical amounts of money, living in the same place, and choosing the same health plan. Who will pay more for that health care plan under Obamacare?&#13;<br /> &#13;<br /> No one would dispute that 30-year-olds have much lower health care costs than 64-year-olds, on average. A compelling illustration comes from a highly cited article <a href="http://qje.oxfordjournals.org/content/111/2/391.short">using the National Medical Expenditure Survey</a>; the authors show that health care costs for 64-year-old women and men are approximately 2 to 4 times that their 30-year-old counterparts (Cutler and Gruber, 1996, p. 429). A natural implication is that groups with higher expected medical costs (such as older individuals), will tend to face higher health care premiums than those with lower expected medical costs (such as younger individuals).&#13;<br /> &#13;<br /> A quick quiz:&#13;<br /> &#13;</p> <ul><li>First, take a 30-year-old and 64-year-old living in Florence, Wisconsin, each of whom is purchasing health insurance on the federal exchange. If they have the same income of $41,000, are non-smokers, and choose the exact same plan, which one faces higher premiums ignoring subsidies?</li> <li>Second, who pays more out of their own pocket, once Obamacare subsidies are included?</li> <p>&#13; </p> </ul><p>The answer to the first question is easy. The 64-year-old faces higher premiums. For example, if the 64-year-old purchased the <a href="https://www.healthcare.gov/see-plans/54121/details/52697WI0010003/?county=55037&amp;data=2df008c36bc2542bc4df584e1d6256632943227bc137536c82af88b88224fb232f1719ef1a02a50c90afc938b47d0f34e43724f8c3a97d6577f05b9fc57a8137d77b093a79b032c1f1bccf79691e4dedac7c703ea4f4019954bd54fa80370b0317e610683977aaf8f0a0873c3a8ce8eb32212fae&amp;source=bitly&amp;state=WI">Molina Marketplace Bronze Plan</a>, he or she would face nearly $7,400 in premiums without subsidies. A 30-year-old would pay around $2,800 for the <a href="https://www.healthcare.gov/see-plans/54121/details/52697WI0010003/?county=55037&amp;data=7587279bce476062d7418b3ca44edc2ee1678b1969370555b14a745b01d65319365d64f97398b39ba62fed167612bb0f7ab62394c123a0c9585f3728e8cee0faff5801d3494c4c88eb079dbb797c881f07856b09787ad25f143a3d9d9008b1d46827d0c32a5b4e1345d9630896a0e0b3ee69cacf&amp;source=bitly&amp;state=WI">same plan</a>.&#13;<br /> &#13;<br /> The second question’s answer may surprise you. Obamacare gives large subsidies for people with incomes between 100% and 400% of the federal poverty line. The more you make, the more you pay for a given plan. But Obamacare gives subsidies that are pegged to a generous benchmark plan (the second lowest cost silver plan). Costs, premiums and subsidies will all be higher for the generous benchmark plan for an older individual than a younger one. However, the subsidy amount can be applied to less generous plans, and this can lead to the surprising result that an older person actually pays less out of their own pocket.&#13;<br /> &#13;</p> <p>Let’s see this in action. If the 64-year-old, non-smoker, making $41,000, purchases a bronze health care plan, he or she pays zero premiums out-of-pocket, after subsidies:&#13;<br /> &#13;<br /><strong>Single Individual, Age 64, Earning $41,000, Florence, WI</strong> <strong>Pays $0 per year for premiums for health plan</strong>&#13;<br /> &#13;<br /><strong></strong></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="87ca9a23-ff94-4cbe-849f-2a5974a77bdc" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/age64_premium_of_0.jpg?itok=fn-SxrFz 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/age64_premium_of_0.jpg?itok=Uc_0Z6QN 1.5x" width="700" height="437" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/age64_premium_of_0.jpg?itok=fn-SxrFz" alt="Media Name: age64_premium_of_0.jpg" typeof="Image" class="component-image" /></p></div> <p>&#13;<br /> &#13;<br /> The subsidies – as noted <a href="https://www.cato.org/blog/obamacare-giveaway-wisconsin-edition-earn-62k-get-free-insurance">here</a>– are very generous; this individual, who makes $41,000, pays nothing for premiums. Obamacare pays 100% of the premium cost. Now let’s examine a 30-year-old in an identical situation:&#13;<br /> &#13;<br /><strong>Single Individual, Age 30, Earning $41,000, Florence, WI</strong> <strong>Pays $2,424 per year for premiums for health plan</strong>&#13;<br /> &#13;<br /><strong></strong></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="755eecf9-b72e-4827-8e1f-1c8adc6ac43a" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/age30_premium_of_2424.jpg?itok=MyFTUuwM 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/age30_premium_of_2424.jpg?itok=ojFVk1wz 1.5x" width="700" height="449" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/age30_premium_of_2424.jpg?itok=MyFTUuwM" alt="Media Name: age30_premium_of_2424.jpg" typeof="Image" class="component-image" /></p></div> <p>&#13;<br /> &#13;<br /> To repeat, everything is identical. Yet, because of the way in which the Obamacare subsidy is pegged to the benchmark plan, the 30-year-old pays thousands of dollars more for the same bronze health plan. Obamacare pays 15% of the premium cost.&#13;<br /> &#13;<br /> This surprising finding won’t be true for all examples, but does illustrate something perverse. In some instances, the subsidies not only lower out of pocket costs for premiums, but completely reverse the logical ordering of who pays higher amounts for a given plan. However, there’s a straightforward fix, one that many private companies and even public institutions practice for their workers: they peg the premium voucher to the most efficient/lowest-cost plan with respect to premiums, and allow workers to apply that voucher to more generous plans.</p> Tue, 07 Jul 2015 09:57:00 -0400 Aaron Yelowitz https://www.cato.org/blog/obamacare-giveaway-its-better-be-64-30-sometimes The Obamacare Giveaway, Wisconsin Edition: Earn $62k and Get Free Insurance https://www.cato.org/blog/obamacare-giveaway-wisconsin-edition-earn-62k-get-free-insurance Aaron Yelowitz <p>The <em>King v. Burwell</em> decision last month highlighted the role of the premium tax credit (i.e. “subsidies”) in Obamacare. I have examined the structure of the subsidies quite carefully, and was shocked by their size. I’ll try to educate you about this, state‐​by‐​state. Today, I’m starting with Wisconsin, which has some of the largest giveaways on the federal exchange. <br /><br /><br /> Among all individuals aged 55 to 64 in Wisconsin, approximately 70% – or <a href="http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_13_3YR_B27004&amp;prodType=table">522,000 people</a> – rely on employer coverage, where the odds are high that they’re paying something out of their own pocket for monthly premiums. <br /><br /><br /> Consider either a single person earning $41,000, or a married couple earning $62,500. Each is 64‐​years‐​old, a non‐​smoker, and lives in Florence, Wisconsin (ZIP code 54121). The structure of Obamacare subsidies means that many individuals who are not poor can find health plans with such large subsidies that they pay absolutely nothing for premiums out of their own pocket. In this case, the <a href="http://go.hc.gov/1RgfaOI">married couple</a> or <a href="http://go.hc.gov/1G34GGJ">single person</a> would qualify for the Molina Marketplace Bronze Plan with zero monthly premium. <br /><br /><br /> See the graphic below for a married couple: <br /></p> <p><strong>Married Couple, Earn $62,500, Florence, WI <br /> Pay $0 per year for premiums</strong></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="591664b4-eaef-4bf8-b4c7-05596707387d" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/20507_blog_yelowitz61.jpg?itok=6DVfGbUT 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/20507_blog_yelowitz61.jpg?itok=zGEwbk5W 1.5x" width="700" height="460" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/20507_blog_yelowitz61.jpg?itok=6DVfGbUT" alt="Media Name: 20507_blog_yelowitz61.jpg" typeof="Image" class="component-image" /></p></div> <p>Although plenty of other plans exist – with higher premiums but less cost sharing– this plan is essentially a giveaway for those who didn’t want to purchase coverage, but were mandated to do so by the government. And if a near‐​elderly person (in the age range of 55 to 64) happens to get sick, they could always move into a plan with generous cost sharing provisions in future years, a problem that economist Martin Feldstein calls Obamacare’s <a href="http://www.project-syndicate.org/commentary/martin-feldstein-on-how-america-s-health-care-reform-could-unravel">fatal flaw</a>.</p> Mon, 06 Jul 2015 12:39:43 -0400 Aaron Yelowitz https://www.cato.org/blog/obamacare-giveaway-wisconsin-edition-earn-62k-get-free-insurance One Consequence of King v. Burwell https://www.cato.org/blog/one-consequence-king-v-burwell Aaron Yelowitz <p>Some say that today’s decision maintains <a href="https://www.cbiz.com/insights-resources/details/articleid/2744/hrb-111-supreme-court-opinion-king-v-burwell-article">business as usual</a> for Obamacare, taxpayers and consumers. The Supreme Court upheld the subsidies (also known as the Premium Tax Credit) to consumers in the 34 states that rely on the federal exchange. Proponents of these subsidies argue they help keep health insurance affordable. <br><br /> <br> The subsidies lower the out‐​of‐​pocket cost to consumers who get them, but at what cost? Consider a&nbsp;<a href="https://www.healthcare.gov/see-plans/33012/results/?data=f518ccf5d992f5f30130bcbe46a2d60a8473c9d262516f081bddc8ba3342e1f989d136d51440be6d9290c7ebf808e0b47458b35ab257a285c5bbe9fa768d7b4aed0a93829dc58ef2bb19555770a8bd7463027fc97efe6fd833e5c4eb5bbebc2cd1cb41d8cbcd5b536a9a395f880e3ba9392f4a21&amp;state=FL&amp;county=12086">64‐​year‐​old consumer in Hialeah, Florida</a> (one of the largest areas impacted by <em>King v. Burwell</em>) who is receiving the maximum subsidy of $7,488 per year. Of the 87 plans offered in the marketplace, 16 entail zero cost to the consumer. Premiums for these “free” plans range from $6,300 to $7,200. There is no incentive for the consumer to shop prudently from these 16 plans. The consumer does not get to keep any unused subsidy, creating incentives to choose health plans with additional features of only marginal value. The taxpayer – not the consumer – picks up the cost of the imprudent choices. <br><br /> <br> In addition to discouraging shopping based on plan value, the premium tax credit offers a&nbsp;set of perverse incentives, especially on the decision to <a href="https://www.cato.org/blog/economic-consequences-aca-notch">earn more than 400% of the poverty line</a> and on <a href="https://www.cato.org/blog/obamacare-reporting-loophole">reporting your income for the upcoming year</a>. <br><br /> <br> Today’s decision may very well mean business as usual, but there are serious economic issues with how the subsidy is set up.</p> Thu, 25 Jun 2015 15:19:11 -0400 Aaron Yelowitz https://www.cato.org/blog/one-consequence-king-v-burwell The Obamacare Reporting Loophole https://www.cato.org/blog/obamacare-reporting-loophole Aaron Yelowitz <p>In many areas of the tax system, loopholes create horizontal inequity in that two nearly identical people pay very different taxes for trivial differences in behavior. Tax schemes for the financially sophisticated abound, such as <a href="https://turbotax.intuit.com/tax-tools/tax-tips/General-Tax-Tips/5-Hidden-Ways-to-Boost-Your-Tax-Refund/INF22336.html">paying mortgages early</a>, <a href="http://www.wsj.com/articles/SB10001424052970204612504574193480955034164">converting 401k funds</a>, and even <a href="http://krugman.blogs.nytimes.com/2009/12/16/throwing-momma-from-the-train/">dodging death tax</a>es. <br /><br /><br /> Obamacare provides a particularly egregious loophole for reporting income. It is a very lucrative yet an unintended scheme. Despite <a href="http://www.politico.com/story/2013/12/irs-obamacare-fraud-tax-credits-100624.html">Sen. Orrin Hatch</a> calling it a “fraudster’s dream come true” back in 2013, the loophole still exists today. <br /><br /><br /> To illustrate the Obamacare reporting loophole, consider the health insurance marketplace in Hialeah, Florida with two consumers. The first, Michael, is single, age 49, a non‐​smoker, and makes $46,000. The second, Lisa, makes $47,000 but is otherwise similar. Both find themselves ineligible for a taxpayer subsidy on <a href="https://www.healthcare.gov/see-plans/33012/?state=FL">Health​Care​.gov</a> and in searching more than 80 plans decide on a <a href="https://www.healthcare.gov/see-plans/33012/details/35783FL1160036/?county=12086&amp;data=8d5a9e141c6997e8777164df86544b893399a87a6a8318fe3ca30267b59b6b71494e3b4c844c1d69674e99efbe4bf82b3b337cb8e2cd6d8b5b1c1e02415f8d896b56da4bb23f48edd78cfc4c5a6f97cd4a0b11a6c8db2d01725f8cb9dc3142e52862a27396d8ea1cf963d4828b93dbc18888125b&amp;start=10&amp;state=FL">Humana Bronze plan</a> with an annual premium of $4,092. <br /><br /><br /> Where’s the reporting loophole? If Michael reports that he expects to make just $12,000 during 2015, he’ll ultimately pay $1,250 for his health insurance. If Lisa does the same, she’ll be on the hook for full amount. The Obamacare reporting loophole lowers Michael’s payment by more than $2,800, even though he wasn’t eligible for a taxpayer subsidy at all. <br /><br /><br /> How does Michael profit from this? Obamacare offers sizable <a href="http://www.irs.gov/publications/p974/">taxpayer subsidies</a> to those with low income. Even so, many would have difficulty paying more than $4,000 in advance for health insurance. Instead, consumers can report their anticipated income and then have the <a href="http://www.irs.gov/Affordable-Care-Act/Individuals-and-Families/Questions-and-Answers-on-the-Premium-Tax-Credit">subsidy advanced directly to the insurance company</a>. Advanced reporting of income runs into a practical issue: Michael or Lisa might make an inaccurate report. If so, the advance subsidy would be incorrect. One might expect that Michael or Lisa would have to square up during tax filing season, a process the IRS calls <a href="http://www.irs.gov/Affordable-Care-Act/Individuals-and-Families/Questions-and-Answers-on-the-Premium-Tax-Credit">reconciling</a>. For single individuals like Michael with income under $46,680 (400% of the poverty line), the way in which the advance subsidy is reconciled encourages misreporting. Michael faces a <a href="http://apps.irs.gov/app/vita/content/02/02_04_010.jsp?level=basic">repayment limit</a> of at most $1,250, if the taxpayer advance to the insurance company was too large. In contrast, there is no upper limit on repayment for Lisa, because her income is above 400% of the poverty line. <br /><br /><br /> The graph below shows how Michael or Lisa profit purely from misreporting, as income changes in relation to the poverty line. In technical terms, if the unsubsidized cost of the second lowest‐​cost silver plan for an individual exceeds $370 per month in 2015, the Obamacare reporting loophole can lead to misreporting subsidies of nearly $3,000. Similar incentives exist for married couples, but with different thresholds and amounts. <br /></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="30175d42-5c01-454b-b494-991a4ff5f3b7" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/201506_yelowitz_blog18.jpg?itok=kogzU6WI 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/201506_yelowitz_blog18.jpg?itok=QJR6LPSe 1.5x" width="700" height="527" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/201506_yelowitz_blog18.jpg?itok=kogzU6WI" alt="Media Name: 201506_yelowitz_blog18.jpg" typeof="Image" class="component-image" /></p></div> <p>What can be done about this loophole? Misreporting arises because repayments are capped. Those above 400% of the poverty line have no incentive to misreport because they would have to fully repay the advance. Aligning incentives to report with actual income by uncapping repayments, as is done for those over 400% of poverty, would remove this loophole. As consumers, financial advisors, and healthcare navigators learn about the Obamacare misreporting loophole, it will be tempting for many people to abuse it, ultimately harming taxpayers.</p> Thu, 18 Jun 2015 11:46:20 -0400 Aaron Yelowitz https://www.cato.org/blog/obamacare-reporting-loophole The Economic Consequences of the ACA Notch https://www.cato.org/blog/economic-consequences-aca-notch Aaron Yelowitz <p>There is great interest in how the labor market will respond to the Affordable Care Act (ACA). Much of the <a href="http://www.wsj.com/articles/unemployed-by-obamacare-1408664211">popular discussion</a> focuses on the implications of the newly‐​implemented and widely‐​anticipated employer mandate, which requires firms with 50 or more workers to provide health insurance for full‐​time employees (defined as workers with 30 or more hours per week). The employer mandate, unsurprisingly, creates strong incentives for companies to scale back employee hours (“29 hour work weeks”) and lay off workers or consolidate part‐​time jobs into full‐​time jobs in order to get under the 50 employee threshold. <br /><br /><br /> There is comparatively less discussion of the incentives faced by workers. Although the <a href="http://www.cbo.gov/sites/default/files/cbofiles/attachments/45010-breakout-AppendixC.pdf">Congressional Budget Office</a> has provided estimates and discussion of the pertinent labor market effects, one issue that tends to get lost in all of this is how increasing a household’s income creates certain “notches” in a household’s budget constraint. By “notches”, economists mean very large changes in the subsidy (known as the “<a href="http://www.irs.gov/publications/p974/">Premium Tax Credit</a>”) received by a household for extremely small changes in income. These notches are well known in other transfer programs, particularly the “<a href="http://yelowitz.com/YelowitzQJE1995.pdf">Medicaid notch</a>” and the “<a href="http://gatton.uky.edu/faculty/yelowitz/Yelowitz-ph.pdf">public housing notch</a>”. The ACA notch occurs in both states that expanded their Medicaid program, as well as those that didn’t. <br /><br /><br /> To illustrate the sheer magnitude of the ACA notch, it is helpful to examine ACA subsidies for different individuals. First, consider a person who is expensive to insure – a 64‐​year‐​old – in a locality that generally has high insurance premiums. A good example is Clay County, Georgia (where Georgia also didn’t expand its Medicaid program). As the “<a href="https://www.healthcare.gov/see-plans/39851/?state=GA">Plan Preview and Price Estimator</a>” from the federal government’s exchange shows, the premium tax credit goes up dramatically for this individual at an income of $11,671 and falls dramatically at an income of $46,679. <br /></p> <p> <strong>Now You See It, Now You Don’t</strong>&#13;<br /><em>The premium tax credit appears when income reaches 100% FPL</em>&#13;<br /></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="1c4f94a6-c177-4692-9dff-6bd38e55a82e" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc11670.jpg?itok=DxA5xa9_ 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/zip39851_inc11670.jpg?itok=5c-_oy2D 1.5x" width="700" height="654" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc11670.jpg?itok=DxA5xa9_" alt="Media Name: zip39851_inc11670.jpg" typeof="Image" class="component-image" /></p></div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="23b4dfea-2580-4dce-925a-9dac13b5e613" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc11671.jpg?itok=NKhOvz7W 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/zip39851_inc11671.jpg?itok=mtSDcooy 1.5x" width="700" height="658" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc11671.jpg?itok=NKhOvz7W" alt="Media Name: zip39851_inc11671.jpg" typeof="Image" class="component-image" /></p></div> <p><em>The premium tax credit disappears when income reaches 400% FPL</em>&#13;<br /></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="3d0047d5-0028-4d77-a5f5-0d9f2f1eaeea" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc46679.jpg?itok=i06_9zXB 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/zip39851_inc46679.jpg?itok=qrnRPfr7 1.5x" width="700" height="662" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc46679.jpg?itok=i06_9zXB" alt="Media Name: zip39851_inc46679.jpg" typeof="Image" class="component-image" /></p></div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="d01e2a4a-9c17-4954-b880-709ea9666fc9" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc46680.jpg?itok=23PU2-BI 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/zip39851_inc46680.jpg?itok=uKYCu80q 1.5x" width="700" height="661" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/zip39851_inc46680.jpg?itok=23PU2-BI" alt="Media Name: zip39851_inc46680.jpg" typeof="Image" class="component-image" /></p></div> <p>What’s going on? Subsides – discounts off the premiums for health plans offered on the exchange (known as the premium tax credit or “PTC”) – are related to household income as well as cost factors (namely an individual’s age and price of health plans in the local marketplace). Subsidies kick in at 100% of the Federal poverty line – or $11,671 for a one‐​person household – and turn off at 400% of the Federal poverty line – or $46,679. Thus, small changes in income lead can lead to very large changes in the subsidy. <br /><br /><br /> Before discussing the labor market consequences, it is important to note that such ACA notches are more important for expensive‐​to‐​insure individuals and couples, and the size of the ACA notch also varies by location. The following table shows a high‐​cost individual (the 64‐​year‐​old) and a low‐​cost individual (a 30‐​year‐​old) in a high‐​cost location (Clay County, GA) and a lower‐​cost location (Andersen County, TN). <br /></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="ebd69d60-716b-42f1-9440-0c3ba4e0011e" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/aca_notch_ga.png?itok=ltzWk4ws 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/aca_notch_ga.png?itok=pWFd5cId 1.5x" width="666" height="385" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/aca_notch_ga.png?itok=ltzWk4ws" alt="Media Name: aca_notch_ga.png" typeof="Image" class="component-image" /></p></div> <p>Sources: <a href="https://www.healthcare.gov/see-plans/39851/?state=GA">https://​www​.health​care​.gov/​s​e​e​-​p​l​a​n​s​/​3​9​8​5​1​/​?​s​t​a​te=GA</a> and <a href="https://www.healthcare.gov/see-plans/37705/?state=TN">https://​www​.health​care​.gov/​s​e​e​-​p​l​a​n​s​/​3​7​7​0​5​/​?​s​t​a​te=TN</a> (Accessed 6/11/2015). <br /><br /><br /> There are several things to take away from this table. First, Georgia and Tennessee are among the <a href="http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/">21 states</a> that have not expanded their Medicaid program. The ACA only provides subsidies for individuals at or above 100% of the Federal poverty line; in states that expanded Medicaid, individuals below 138% of the Federal poverty line would qualify for Medicaid. Second, for the 64‐​year‐​old, the first ACA notch – in states without a Medicaid expansion – creates dramatic subsidies once income reaches 100% of the Federal poverty line, or $11,671. Earning the extra $1 after $11,670 raises the subsidy by $10,849 per year in Clay County, GA, but only $5,910 in Andersen County, TN. Both of these ACA notches – which wouldn’t be present in the Medicaid expansion states – create strong incentives to increase work effort to reach this threshold. As can also be seen, the ACA notches are present but less dramatic for the younger person. Third, there are “mini ACA notches” as income exceeds certain multiples of the Federal poverty line. As the 64‐​year‐​old individual earns the extra $1 in Georgia that raises income from $15,521 to $15,522 (133% of the Federal poverty line), the subsidy falls by $157. Fourth, once income exceeds 400% of the Federal poverty line, the subsidy disappears entirely. For this individual, that entails a loss of subsidy of $6,621 from earning the extra $1 that takes income from $46,679 to $46,680. This notch is also present in Tennessee, but to a smaller extent. Finally, in all cases we can see the subsidy typically erodes quite smoothly as income goes up – this is known as a benefit reduction rate or tax rate. As income increases by $33,000 from $12,000 to $45,000, the PTC falls by $4,061, resulting in an average tax rate of 12.3% just from the ACA. For the younger individual, the subsidy erodes to $0 before income reaches 400% of the Federal poverty line in both Georgia and Tennessee. <br /><br /><br /> How do things look for married couples? Much like single individuals, the subsidies kick in and turn off at multiples of the Federal poverty line. Although the unsubsidized cost of a health insurance plan for two 64‐​year‐​olds is twice that of one 64‐​year‐​old, the dollar amounts for the poverty thresholds are quite different. The dollar amounts go up less than proportionally with family size. As a consequence, the notches look quite different – and in some cases are jaw‐​dropping – for a married couple. Consider the two areas we just considered, and assume that two individuals of the same age are married to each other. The first column in the next table shows that the ACA notch when reaching 100% of the Federal poverty line (of $15,731) is an incredible $21,850! That is, earning the extra $1 that brings income from $15,730 to $15,731 leads to a dramatic increase in the premium tax credit. The magnitudes are clearly different, but present, for all family types illustrated. As family income goes from $15,731 to $62,919 (or 100% to 400% of the Federal poverty line), for all couples, the subsidy more‐​or‐​less is smoothly taxed away (and in, fact, the young couple in the inexpensive market loses its subsidy before 400% of the Federal poverty line). For the first couple, as income goes from $18,000 to $60,000, the PTC falls by $5,374, resulting in an average tax rate from the ACA alone of 12.8%. The notch for older couples is dramatic at 400% of the Federal poverty line; in Clay County, GA, earning the extra $1 that takes income from $62,919 to $62,920 results in a loss of subsidy of $16,152! The results in Tennessee are also large, but not nearly as large as Georgia. In Tennessee, the older couple only loses $6,275 for earning the extra $1. Younger couples don’t completely escape this punitive tax. For younger couples, the ACA notch exists in Georgia, but the PTC is eroded completely in Tennessee before income reaches 400% of the Federal poverty line, so there is no ACA notch. <br /></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="137f4f3d-60d4-4d6c-be86-ecb36c11d7a5" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.01_am.png?itok=YoSynPIW 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.01_am.png?itok=JeL5YGD0 1.5x" width="655" height="443" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.01_am.png?itok=YoSynPIW" alt="Media Name: screen_shot_2015-06-16_at_10.22.01_am.png" typeof="Image" class="component-image" /></p></div> <p>Sources: <a href="https://www.healthcare.gov/see-plans/39851/?state=GA">https://​www​.health​care​.gov/​s​e​e​-​p​l​a​n​s​/​3​9​8​5​1​/​?​s​t​a​te=GA</a> and <a href="https://www.healthcare.gov/see-plans/37705/?state=TN">https://​www​.health​care​.gov/​s​e​e​-​p​l​a​n​s​/​3​7​7​0​5​/​?​s​t​a​te=TN</a> (Accessed 6/11/2015). <br /><br /><br /> How would such incentives affect the labor market? Abstracting away from other taxes and transfers, these notches create incentives in all cases to reach the earnings threshold of 100% of the Federal poverty line in order to qualify for subsidized health insurance. Moreover, there are very strong incentives to not exceed 400% of the Federal poverty line, especially because <a href="http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/">you must repay all of the premium tax credit. </a>In states that did not expand Medicaid, the first effect – the incentive to raise earnings above 100% of the Federal poverty line – is present, but isn’t in states that expanded Medicaid. In all 50 states and DC, the second ACA notch at 400% of the Federal poverty line will be present, to larger or smaller degrees depending on health premiums and age. The larger the ACA notch, the greater the incentive to constrain earnings under the second threshold. <br /><br /><br /> It is also the case that this structure creates unusual marriage taxes and bonuses, an incentive that has been examined in the context of <a href="http://yelowitz.com/YelowitzJHR1998.pdf">Medicaid expansions</a> from an earlier era. To illustrate, imagine that two unmarried, 64‐​year‐​olds in Clay County, GA each had annual income of $10,500. The first table illustrates that neither would be eligible for the PTC. By marrying, household income is $21,000, resulting in a premium tax credit of $21,526. However, not all couples look so good. Consider these same two individuals, each earning $33,000. As single individuals, they each receive a premium tax credit of $8,094, or a cumulative amount of $16,188. By marrying, their credit would fall to $0, because household income would exceed the limit of 400% of the Federal poverty line. Evidence from the ACA mandate to cover young adults shows that <a href="https://appam.confex.com/appam/2014/webprogram/Paper10104.html">marriage taxes and bonuses</a> are an important factor. <br /><br /><br /> Graphical Summary <br /></p> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="1011e847-32b0-460b-9229-5bd50fce4bfc" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.18_am.png?itok=_xr1IZ8A 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.18_am.png?itok=6DhCTQ_i 1.5x" width="565" height="345" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.18_am.png?itok=_xr1IZ8A" alt="Media Name: screen_shot_2015-06-16_at_10.22.18_am.png" typeof="Image" class="component-image" /></p></div> <div data-embed-button="image" data-entity-embed-display="view_mode:media.blog_post" data-entity-type="media" data-entity-uuid="fb47c0b2-43fd-4413-bbee-81a74bf5a257" data-langcode="und" class="embedded-entity"> <p><img srcset="/sites/cato.org/files/styles/pubs/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.29_am.png?itok=imoGBS30 1x, /sites/cato.org/files/styles/pubs_2x/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.29_am.png?itok=MaaXNiV5 1.5x" width="590" height="345" src="https://www.cato.org/sites/cato.org/files/styles/pubs/public/wp-content/uploads/screen_shot_2015-06-16_at_10.22.29_am.png?itok=imoGBS30" alt="Media Name: screen_shot_2015-06-16_at_10.22.29_am.png" typeof="Image" class="component-image" /></p></div> <p>Cited Work: <br /> Wall Street Journal, “Unemployed by Obamacare,” August 21, 2014, Accessed from: <a href="http://www.wsj.com/articles/unemployed-by-obamacare-1408664211&amp;#13">http://​www​.wsj​.com/​a​r​t​i​c​l​e​s​/​u​n​e​m​p​l​o​y​e​d​-​b​y​-​o​b​a​m​a​c​a​r​e​-​1​4​0​8​6​6​4​2​1​1&amp;#13</a>;<br /><br /><br /> Congressional Budget Office, “The Labor Market Effects of the Affordable Care Act,” February 2014, Accessed from: <a href="http://www.cbo.gov/sites/default/files/cbofiles/attachments/45010-breakout-AppendixC.pdf&amp;#13">http://​www​.cbo​.gov/​s​i​t​e​s​/​d​e​f​a​u​l​t​/​f​i​l​e​s​/​c​b​o​f​i​l​e​s​/​a​t​t​a​c​h​m​e​n​t​s​/​4​5​0​1​0​-​break…</a>;<br /><br /><br /> Internal Revenue Service, Publication 974: The Premium Tax Credit, March 2015, Accessed from: <a href="http://www.irs.gov/publications/p974/&amp;#13">http://​www​.irs​.gov/​p​u​b​l​i​c​a​t​i​o​n​s​/​p​9​7​4​/&amp;#13</a>;<br /><br /><br /> Yelowitz, A., “The Medicaid Notch, Labor Supply and Welfare Participation: Evidence from Eligibility Expansions,” The Quarterly Journal of Economics, November 1995, 110(4): 909–939. <br /><br /><br /> Yelowitz, A., “Public Housing and Labor Supply,” Mimeo, University of Kentucky, November 2001. <br /><br /><br /> Kaiser Family Foundation, “Status of State Action on the Medicaid Expansion Decision,” Accessed from: <a href="http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/&amp;#13">http://​kff​.org/​h​e​a​l​t​h​-​r​e​f​o​r​m​/​s​t​a​t​e​-​i​n​d​i​c​a​t​o​r​/​s​t​a​t​e​-​a​c​t​i​v​i​t​y​-​a​r​o​u​n​d​-expa…</a>;<br /><br /><br /> Yelowitz, A., “Will Extending Medicaid to Two Parent Families Encourage Marriage?” The Journal of Human Resources, Fall 1998, 33(4): 833–865. <br /><br /><br /> Abramowitz, J., “Saying ‘I Don’t’: The Effect of the Affordable Care Act Young Adult Provision on Marriage,” Accessed from: <a href="https://appam.confex.com/appam/2014/webprogram/Paper10104.html">https://​appam​.con​fex​.com/​a​p​p​a​m​/​2​0​1​4​/​w​e​b​p​r​o​g​r​a​m​/​P​a​p​e​r​1​0​1​0​4​.html</a></p> Tue, 16 Jun 2015 11:52:00 -0400 Aaron Yelowitz https://www.cato.org/blog/economic-consequences-aca-notch The Obamacare Earnings Cliff https://www.cato.org/multimedia/daily-podcast/obamacare-earnings-cliff Aaron Yelowitz <p>The incentive structure built into Obamacare create earnings cliffs that may alter the behavior of millions of Americans. Aaron Yelowitz explains the problem.</p> Fri, 12 Jun 2015 17:30:00 -0400 Aaron Yelowitz https://www.cato.org/multimedia/daily-podcast/obamacare-earnings-cliff Aaron Yelowitz testifies on the proposed minimum wage ordinance at the Louisville City Council https://www.cato.org/multimedia/media-highlights-tv/aaron-yelowitz-testifies-proposed-minimum-wage-ordinance-louisville Thu, 30 Oct 2014 12:12:00 -0400 Aaron Yelowitz https://www.cato.org/multimedia/media-highlights-tv/aaron-yelowitz-testifies-proposed-minimum-wage-ordinance-louisville ObamaCare: Still a Bad Deal for Young Adults https://www.cato.org/blog/obamacare-still-bad-deal-young-adults Aaron Yelowitz <p>The Associated Press reports that <a href="http://www.google.com/hostednews/ap/article/ALeqM5hLAMW_KTqY_JVMQF-gNn3O0_uUcQD9EOIBQO0">young adults will face higher premiums under ObamaCare</a>: </p> <blockquote><p>Beginning in 2014, most Americans will be required to buy insurance or pay a&nbsp;tax penalty. That’s when <strong>premiums for young adults seeking coverage on the individual market would likely climb by 17 percent on average, or roughly $42 a&nbsp;month</strong>, according to an analysis of the plan conducted for The Associated Press. The analysis did not factor in tax credits to help offset the increase. <br><br /> <br> The higher costs will pinch many people in their 20s and early 30s who are struggling to start or advance their careers with the highest unemployment rate in 26&nbsp;years.</p> </blockquote> <p>The article cited additional studies estimating premiums increases young adults as high as 50 percent. That was essentially the message of my recent Cato briefing paper, “<a href="https://www.cato.org/pub_display.php?pub_id=10933">ObamaCare: A&nbsp;Bad Deal for Young Adults</a>.” <br><br /> <br> Supporters claim the new law provides subsidies that would help people afford the higher premiums. As I&nbsp;write in my paper, however: </p> <blockquote><p><strong>The money for those subsidies has to come from somewhere</strong>, though. Presumably, some of it would come from young adults themselves in the form of higher taxes or the tax penalties imposed on those who do not purchase insurance…So<strong> the presence of subsidies does not necessarily mean that young adults would come out winners. </strong>Ironically, all the complexity may actually help the legislation pass Congress precisely because it obscures whom the legislation would tax.</p> </blockquote> <p>Supporters also claim that although the higher premiums might be actuarially unfair for people who are young and healthy today, those people will eventually be old and unhealthy. Over the course of a&nbsp;lifetime, they reason, such policies would be closer to actuarially fair. <br><br /> <br> The problem is that we’ve heard this line before. Inter‐​generational redistribution is fundamentally unfair to the young because it creates a&nbsp;situation where the old, who vote, have incentives to ratchet up benefits – and to ratchet up taxes on the young, who don’t vote. Social Security collects from the young and gives to the old, and is clearly a&nbsp;net tax on the young. As Jonathan Gruber <a href="http://books.google.com/books?id=phgD_KvT06UC&amp;pg=PA337&amp;lpg=PA337&amp;dq=%E2%80%9CYoung+Americans+and+Social+Security&amp;source=bl&amp;ots=vEGoFy5Fpt&amp;sig=yTJOmMMbMI_jK-nBR_vIqNN3Rm4&amp;hl=en&amp;ei=JxCyS8S4I5WQNqu7-MAK&amp;sa=X&amp;oi=book_result&amp;ct=result&amp;resnum=4&amp;ved=0CBAQ6AEwAw#v=onepage&amp;q=%E2%80%9CYoung%20Americans%20and%20Social%20Security&amp;f=false">reports</a>, the young have very little confidence – deservedly so – in Social Security’s implicit promises. Experience shows that whatever new taxes ObamaCare imposes on the young will grow over time. <br><br /> <br> Regardless, most young uninsured people already obtain insurance as they get older. As I&nbsp;report in my paper, 30.4 percent of those age 20–29 were uninsured in 2008 (including 33.8 percent of 23‐​year‐​olds), but only 13.4 percent of those aged 50–64&nbsp;years were uninsured. So a&nbsp;significant number of uninsured young adults naturally transition into insured older adults. The main effect of the new law will be to take young adults who think health insurance is a&nbsp;bad deal at today’s prices and force them to health insurance at even higher prices.</p> Tue, 30 Mar 2010 14:48:50 -0400 Aaron Yelowitz https://www.cato.org/blog/obamacare-still-bad-deal-young-adults Obama’s Other Massachusetts Problem https://www.cato.org/publications/commentary/obamas-other-massachusetts-problem Aaron Yelowitz, Michael F. Cannon <div class="lead text-default"> <p>The flood of angry voters in Massachusetts isn’t the only blue‐​state problem coming from Ted Kennedy’s old stomping grounds. </p> </div> , <div class="text-default"> <p>In 2006, Massachusetts enacted a&nbsp;health care law that is essentially identical to the Obama plan — and voters there have seen the consequences. Both Obamacare and the Massachusetts plan aim to expand health insurance coverage via private‐​sector mandates, private‐​insurance subsidies, an expanded Medicaid program and a&nbsp;new health insurance “exchange.” </p> <p>President Obama has avoided any comparisons between his plan and the Massachusetts law, with good reason. Premiums for employer‐​sponsored insurance — 96 percent of the Massachusetts’ market — are rising 21 percent to 46 percent faster than the national average. State officials have whitewashed the cost overruns, but they are simultaneously raising taxes and threatening to impose a&nbsp;Canadian‐​style payment system, in which doctors and hospitals do the dirty work of rationing care. </p> <p>In a&nbsp;new <a href="https://www.cato.org/pub_display.php?pub_id=11115">Cato Institute study</a>, we found that the Massachusetts law’s runaway costs are delivering far less than supporters claim. Using 2008 data from the Census Bureau’s Current Population Survey, “the survey of record” for measuring trends among the uninsured, and controlling for relevant variables and using other New England states to control for unobserved factors, we found unflattering results. </p> <p>Official estimates of coverage gains overstate the law’s impact by at least 45 percent. At best, the law covered 297,000 previously uninsured residents, and the uninsured rate is 3.8 percent rather than 432,000 and 2.6 percent, as Massachusetts claims. </p> <p>Yet even those estimates are overly optimistic, as Massachusetts appears to have driven many uninsured residents underground. </p> <p>There is evidence that many residents responded to the law — whose “individual mandate” makes it a&nbsp;crime not to have health insurance — by concealing their insurance status from the Census Bureau. Our results suggest that Massachusetts’ actual uninsured rate may be 5.1 percent or higher and that state officials may be overstating the coverage gains by 112 percent or more. </p> <p>In addition, the law’s Medicaid expansion spurred many residents to substitute Medicaid for private insurance, much as critics fear a “public option” would. Private coverage fell by 14.6 percentage points among low‐​income children — despite no discernable increase in total coverage — and by 6.2 points among low‐​income adults. (Because the Senate version of the president’s plan would expand Medicaid, maybe it contains a “public option” after all.) </p> <p>We also found that the law has done a&nbsp;better job of giving residents coverage than moving the population toward better health. Self‐​reported health improved for some residents but fell for others. </p> <p>Finally, it appears the law has made Massachusetts a&nbsp;less attractive place to live for young adults. </p> <p>Even before 2006, Massachusetts forced young consumers to pay inflated premiums for the purpose of subsidizing their elders. The law’s new “individual mandate” forces those young adults to accept that bad deal or pay a&nbsp;penalty. As a&nbsp;result, the number of young adults relocating to Massachusetts has fallen by 60 percent. </p> <p>The Obama plan and the Massachusetts law bear another similarity: So far, no one has bothered to estimate their full cost. </p> <p>The nonpartisan Massachusetts Taxpayers Foundation used creative accounting to declare the law’s cost to be “modest.” Using unpublished data provided to us by the foundation, however, we conservatively estimate total new spending under the law to have exceeded $1 billion in 2008 — 57 percent higher than the foundation’s published estimate. The total cost is much higher, as our estimate includes only one category of mandatory private‐​sector spending. </p> <p>But at least we have a&nbsp;vague idea of the full cost of the Massachusetts law. Unlike the way it handled the Clinton health plan, the Congressional Budget Office has produced no cost estimates of the Obama plan’s private‐​sector mandates. A&nbsp;recent CBO memo reveals that Democrats have been working meticulously to suppress any such estimates. </p> <p>Massachusetts has done the nation a&nbsp;great favor. It is providing a&nbsp;preview of life under the Obama health plan: a&nbsp;lot of pain for very little gain. </p> </div> Thu, 21 Jan 2010 00:00:00 -0500 Aaron Yelowitz, Michael F. Cannon https://www.cato.org/publications/commentary/obamas-other-massachusetts-problem