Topic: Regulatory Studies

Do We Need Fannie and Freddie?

All eyes were on Wall Street Monday morning as Freddie Mac, one of the two giant government-sponsored enterprises that dominate U.S. mortgage finance, floated $3 billion in bonds to continue its role as a buyer and reseller of mortgage obligations. The bond sale came after a week in which Freddie’s stock, and the stock of its sister Fannie Mae, plummeted as analysts and economists worried that the housing bubble collapse would push the two GSEs into insolvency. To fortify Fannie and Freddie, the Bush Treasury Department, with blessings from the Democratic Congress, worked feverishly over the weekend to cobble together a bailout plan should the GSEs’ conditions worsen to the point that they can no longer function. The successful bond sale indicates the Bush plan has reassured the market — albeit barely.

It is an article of faith across the political spectrum that Freddie and Fannie cannot be allowed to fail — especially not at this time when the broader housing market is undergoing painful correction. Why do they have such exalted status?

Before Fannie Mae — the first of the twins — was created amidst the “Recession within the Depression” in 1938, home mortgage lending was highly risky for banks. State regulation kept banks small and geographically limited in order to make them better targets for taxation and political manipulation. As a result, banks could not geographically diversify their loan risk, leaving them highly vulnerable to localized economic downturns. Because they lent money (as mortgages and business loans to farms and other firms) to local borrowers for long periods of time but they had to honor local depositors’ withdrawal requests, banks were often one bad harvest and one bank run away from insolvency. For that reason, they shied away from financing long-term home loans.

Fannie Mae (formally, the Federal National Mortgage Association) provided badly needed lubricant to the mortgage industry. It purchased loan obligations from banks, putting money back in the banks’ vaults and making that money available for more home loans. Fannie financed its operations by borrowing on Wall Street and, later, by pioneering the creation and sale of mortgage-back securities (MBSs) — selling large “bundles” of loans to investors and then servicing the loans on the investors’ behalf. In this way, Fannie diversified loan risk, allowing a nationwide (and worldwide) pool of investors to finance (at first indirectly, then directly) a nationwide pool of mortgages. This wasn’t the ideal solution that banking reform would have been, but Fannie was a good second-best solution.

Because of its tax-free status and government backing (as well as banking regulations that constrained would-be competitors), Fannie quickly came to dominate the mortgage industry. This system hummed along, unchanged, until 1968 when Fannie’s costs conflicted with Lyndon Johnson’s efforts to rein in the federal budget amidst the war in Vietnam and the war on poverty. It was decided that Fannie would be spun off as a private corporation whose investors would bear its costs. However, Fannie retained its tax-free status and received a $2.25 billion line of credit at the U.S. Treasury. It also was allowed to operate with much lower reserve requirements than banks. But the most valuable of Fannie’s parting gifts was its implicit too-big-to-fail status: because of its history and role in mortgage lending, investors believe the federal government will ride to Fannie’s aid if it ever became financially unable to function. For this reason, investors buy Fannie’s stocks, bonds, and MBSs.

Congress realized that, with those perks and its original dominant position, Fannie would continue to monopolize the U.S. home loan market. In 1970 they created Freddie Mac (formally, the Federal Home Loan Mortgage Corporation) as a competitor, with the same structure and perks as Fannie. It’s unclear why Congress believed a duopoly was better than a monopoly. (For more on Fannie and Freddie’s history, listen to Peter Van Doren’s recent podcast.)

Over the last decade, analysts have offered numerous criticisms of Fannie and Freddie’s dominance and nature, and issued calls for reform. (In Regulation alone, see: Van Order 2000, Frame & White 2004, Wallison 2004, Jaffe 2006.) Besides the GSEs’ dominance of the market and the risk that their too-big-to-fail status poses to taxpayers, the implied guarantee encouraged the GSEs to retain possession of some of their mortgages instead of selling them to investors. Retaining mortgages and reaping the payments enriched Fannie and Freddie shareholders, but it subjected the two GSEs to interest rate risk as well as the default risk of all Fannie- and Freddie-guaranteed loans.

Despite the criticisms and reform calls, the issue gained little traction, even in the go-go real estate market of the mid-2000s when Freddie and Fannie were involved in a relatively small 40% of new home loans. The result is that, in the aftermath of the housing bubble collapse when preserving mortgage financing is incredibly important, the market for those loans is dominated by two teetering giants.

Fannie and Freddie are currently involved in roughly half of all U.S. mortgages. That includes a large majority of post-bubble mortgages, as private investors have pulled back from financing non-guaranteed loans. Given that position and the overall turmoil in the housing market, the federal government cannot leave Fannie and Freddie to struggle — after all, their debt equals the debt of all other U.S. corporations combined, and imagine what would happen if all U.S. corporations suddenly defaulted on their loan payments.

So, there seems little that federal policymakers can do now except promise to prop up Freddie and Fannie if they become insolvent and hope that doesn’t occur. But this situation demonstrates why the mortgage industry should not be dominated by two firms — especially two government-sponsored firms.

If Congress does adopt legislation to protect Fannie and Freddie, that legislation should spare taxpayers and the nation from a repeat of the current crisis. The legislation should include an ironclad commitment to make Fannie and Freddie fully independent of the government over the next decade. It should also require that the GSEs be broken into several smaller firms that aren’t too big to fail. Those firms would be put under strict government oversight and conservative risk controls until they are fully independent. The legislation should also strip Fannie and Freddie of their tax-free status, putting them on equal footing with private lenders (including banks, which can now geographically diversify following the 1999 banking reform). With this reform implemented, the mortgage market would become supported by numerous institutions (some of which would be Freddie and Fannie spin-offs, others not) instead of just two vulnerable pillars.

It is because policymakers refused to reform state banking regulation in the 1930s that Fannie (and later Freddie) were created. It is because Fannie and Freddie were not reformed in the last 10 years that we’re stuck with this problem now. Federal policymakers cannot disregard another opportunity at reform — or else the U.S. mortgage industry will remain at the mercy of the financial health of two firms.

Postscript (7/15): Cato senior fellow Gerald O’Driscoll sketches a plan for breaking up Fannie and Freddie and making them truly independent in this excellent WSJ op-ed [$].

Dionne’s April Fool’s Column

Anybody can play an April Fool’s Joke in April, but E.J. Dionne deserves credit for pulling a fast one on gullible readers in July. And I’m brave enough to admit that I was briefly fooled by his column asserting that even conservatives now recognize that free markets don’t work. It was only after thinking about his column that I realized he surely must be engaging in some leg pulling if the first person he quotes is one of the most collectivist-minded members of Congress, Barney Frank. Dionne tries to trick readers by then citing the Chairman of the Federal Reserve, who (gee, what a surprise) is in favor of more regulatory power for the Federal Reserve, but he neatly avoids any explanation for why this is evidence that conservatives are abandoning markets (perhaps he is assuming that Bernanke is a conservative because he was appointed by Bush, but surely Dionne is not so naive):

Since the Reagan years, free-market cliches have passed for sophisticated economic analysis. But in the current crisis, these ideas are falling, one by one, as even conservatives recognize that capitalism is ailing. …The old script is in rewrite. “We are in a worldwide crisis now because of excessive deregulation,” Rep. Barney Frank (D-Mass.), the chairman of the House Financial Services Committee, said in an interview. …While Frank is a liberal, the same cannot be said of Ben Bernanke, the chairman of the Federal Reserve. …Bernanke sounded like a born-again New Dealer in calling for “a more robust framework for the prudential supervision of investment banks and other large securities dealers.”

Wait a minute. Perhaps Dionne is writing a serious column. He quotes Irwin Stelzer of the Hudson Institute, who reasonably can be considered a conservative:

What’s striking is that conservatives who revere capitalism are offering their own criticisms of the way the system is working. Irwin Stelzer, director of the Center for Economic Policy Studies at the Hudson Institute, says the subprime crisis arose in part because lenders quickly sold their mortgages to others and bore no risk if the loans went bad. “You have to have the person who’s writing the risk bearing the risk,” he says. “That means a whole host of regulations. There’s no way around that.”

Dionne seems impressed that Stelzer says that markets don’t work perfectly. But that is a reflection of Dionne’s unfamiliarity with economics. After all, failure, like success, is a part of the market process. Dionne does note, however, that Stelzer is endorsing more regulation, so there is a tiny shred of evidence for his hypothesis that conservatives want more government intervention. But if this is the evidentiary bar that has to be cleared for such assertions, I’m going to write a column saying that all socialists now support a flat tax. And I won’t even have to find one left-leaning writer to “prove” my point. I can just point to the various socialist-led governments in Eastern Europe that have adopted single-rate tax systems.

Before signing off, I feel compelled to point out that Stelzer has a fair diagnosis but a misguided prescription. Yes, hindsight shows that lenders were cavalier about loans since they knew other investors would be the ones bearing the risk. But after absorbing billions of dollars in losses, investors obviously have a huge incentive to avoid the same mistake. Indeed, that is why failure plays a crucial role in a market economy; people learn from mistakes. Additional government regulation, by contrast, is at best a case of closing the barn door after the horse has escaped. In the vast majority of cases, however, regulations throw sand in the gears and/or distort incentives for the efficient allocation of resources.

EPA: Your Life is Worthless

Splashed across the front pages today — well, at least one paper I saw — are headlines about the EPA slashing the value of life revising the value of a statistical life downward. This is highly newsworthy, but only because most people haven’t been paying attention to economics or regulatory policy. (One can’t really blame them …)

There are two things that make the news juicy: the fact that regulators are placing a value on life, and the fact that they’re revising the value down.

Most people don’t know that you can put a value on human life. Most people don’t know that they put a value on their own lives all day every day. The slogans that we grow up with - “life is precious” - dominate their thinking. Our parents value our lives very highly and teach us to at least talk about the value of life in exaggerated terms.

This kind of talk and thinking isn’t universal, of course — in our culture and others, sacrificing one’s life for a high ideal is well regarded, as is sacrificing one’s life for science, or for fun. That said, being cavalier or anti-life is generally not a good idea. No, there’s some balance between prizing life and prizing fun, the greater good, ideology, religion, or what-have-you.

We do strike those balances every day. When we go to cross the street, we make judgments about the threat to our life and health from oncoming cars and decide whether to cross in the middle of the block, at a cross-walk, at a controlled intersection, or at a pedestrian footbridge. Most of us have had occassion at least once to think about crossing a freeway — and we haven’t done it.

All this is because we are weighing the value of getting to the other side against the risk of costing ourselves our own lives. To articulate this balancing, what economists are doing is using a dollar value to measure the relative importance of life versus other things.

Think about the alternative: What if you had no way of balancing the value of life against the value of going to the movie theater? People might step into six lanes of onrushing traffic just to be first in the popcorn line. People might cower at the side of an empty two-lane road, passing up a small-town-theater showing of Fun in Acacpulco for fear of setting foot on macadam where a car tire has been. You’ve got to have some measure of the value of life, and you’ve got to use it.

Now, what about the second issue: revising the value of life downward? Under the “life is precious” presumption, that sounds horrible. It should always be revised upwards, right? Well, guess what. If you do, you’re gonna miss the movie.

If you value life too highly, you will take steps to protect life and health that undermine the value of living. Why is life “precious”? Some say for it’s own sake. But most people believe it’s because of the wonderful range of experiences, adventures, tastes, emotions, and relationships we get to enjoy in life. The freedom. If we give up too much of that, focusing strictly on keeping our hearts pumping and air flowing in and out of the lungs, we’ve lost track of the reason for living. Simply maintaining bodies in a state of sentience is not what it’s all about. So regulatory policy must do what we must do as individuals: strike a balance between life and living. Fall too far out of balance in either direction and you’re either prematurely dead or living a life without meaning.

I know nothing about the methodology that the EPA is using to calculate the value of a human life. They came up with $6.9 million. Frankly, that sounds fair to me. (So would $10.2 million, though, or $5.5 million.) There is one problem with it, though. It’s not the value I place on my life. It’s their estimate of the average value that the average American places on his life. Coming up with a single number is a gross collective judgment about how much risk and how much safety each of us should have. It’s incredibly dehumanizing to be lumped together with everyone else this way.

If you disagree with placing a dollar value on human life, well, you disagree with the idea of describing human action in a standardized way. You might as well disagree with giving names to colors.

But if you disagree with the value the EPA is placing on human life, there might be something to that. The regulatory process makes a huge collective judgments about the value of life, lumping us all together into one big average.

We should be as free as possible to make our own judgments about risk and the value of life. It’s difficult with things like air pollution, but even those kinds of risks can often be controlled through individual judgments.

Whatever the case, get over your concerns about placing a dollar value on human life. And revising the value down? — that’s a good thing. It means that we get to have more freedom and more fun!

Nordhaus’s Less-than-optimal Climate Strategy

In “Pointless to rush a carbon emissions plan,” the Toronto Globe and Mail’s Neal Reynolds compares Yale Professor William Nordhaus’s “optimal” approach to controlling greenhouse gases and finds it superior to approaches that would impose deeper controls more rapidly, such as those favored by Stern, various EU leaders, and many in the US.

Under the Nordhaus approach, which is also discussed by Keith Johnson at the Wall Street Journal, costs of control would start at 0.3 per cent of global GDP in 2010 (currently around $60,000 billion), increase to 0.5 per cent in 2015, 0.6 per cent in 2020 and peak at 0.9 per cent in 2065. He estimates the net present value (NPV) of climate change damages absent any controls at $22 trillion. Under this so-called “optimal” approach, the NPV costs of controls would be $2 trillion and climate change damages would be reduced by $5 trillion (i.e., the “optimal” policy would provide net benefits of $3 trillion, but residual damages would be $17 trillion). As he explains, “More of the climate damages are not eliminated because the additional abatement would cost more than the additional reduction in damages.”

He also estimates that proposals that emphasize “excessively early reductions [make] the policies much more expensive… For example, the Gore and Stern proposals have net costs of $17 trillion to $22 trillion relative to no controls; they are more costly than doing nothing today.” By his calculations, his proposal is clearly superior to these other reduction proposals.

However, while Nordhaus’s prescription may indeed be the most “optimal economic approach” to slow global warming, it isn’t the optimal approach to addressing global warming. This is because it ignores adaptation. Some adaptations may reduce climate change damages more efficiently than mitigation. Perhaps all or part of the $2 trillion that Nordhaus would spend on mitigation should, instead, be invested in adaptation. That might reduce damages by more than the $5 trillion. In any case, with adaptation in the mix, $5 trillion may well be the lower bound for the optimal reduction in climate change damages. And, of course, emission reductions that seem to be optimal under 0.9 percent of GDP in 2065 in the absence of adaptation may, once adaptation is thrown into the mix, no longer be optimal.

In fact, a recent Cato Policy Analysis indicates that in the short-to-medium term, adaptation — specifically, reducing vulnerability to climate-sensitive problems that might be exacerbated by climate change — would provide greater benefits than mitigation, and at a much lower cost. Most of those benefits come from the fact that one approach to adaptation is to advance adaptive capacity. Significantly, that can help society cope not only with climate change but, more importantly, to other problems that are more important than climate change now and in the foreseeable future. Thus the ancillary benefits of increasing adaptive capacity are very high, higher than climate change damages in the absence of any controls according to the Cato Policy Analysis.

Notably, Nordhaus acknowledges to having “relatively little confidence in our projections beyond 2050.” To his credit, this skepticism informs his recommended approach, but it would probably have been best to avoid stretching the analysis to 2200.

Sometimes such long-range analyses are justified on the grounds that that’s the best that can be done. But even if that’s so, it misses the real issue, namely, whether even the best available analysis is good enough for making trillion-dollar decisions which, moreover, extend out centuries hence. At these temporal distances, Nostradamus may be just as credible as Nordhaus, or Nicholas Stern, for that matter.

Humility isn’t an offense, and it ought to be acceptable for economists and policy analysts—even those whose stock in trade is climate change—to admit that they haven’t a clue what the world will look like beyond 2050 (if then).

Nordhaus’s numbers indicate that estimates of pre-control damages and post-control residual damages frequently are substantially larger than either the costs or benefits of emission controls. But the treatment of damages (i.e., impacts) of climate change in the Nordhaus analysis is somewhat sketchy. As far as I can determine, none of the damage studies properly account for adaptive capacity, particularly considering that that capacity ought to increase if societies accumulate wealth, human capital and technology at rates implied by all the socioeconomic scenarios used to derive future emissions (and climate change). (See, for example, here.) Thus, both pre-control climate change damages and post- control residual damages could be substantially overestimated.

[Some argue that they disbelieve that economic growth will be as high as assumed, but in that case they should also disbelieve estimates of future climate change and impacts predicated on that growth.]

To summarize, the Nordhaus analysis probably overestimates climate change damages. In any case, the Nordhaus approach could be made more optimal by adding to it an adaptation component that would enhance societies’ adaptive capacities (by reducing present day vulnerabilities to climate-sensitive problems and boosting economic development and human capital in developing countries). In fact, optimal carbon taxes (or cap-and-trade approaches) can only be determined after completion of more comprehensive analyses that include full and equal consideration of adaptation and any ancillary (net) benefits.

Of course that still leaves the problem of relying on analyses over time frames that demand, in Coleridge’s words, “willing suspension of disbelief.” Instead of suspending disbelief and succumbing to gullibility, I would recommend a somewhat different approach (see here, p. 37).

They Want to Post WHAT?

According to the Raleigh News & Observer:

The [North Carolina medical] board, charged with licensing and disciplining the 22,000 doctors who practice in North Carolina, has proposed posting all malpractice payments going back seven years as part of a new effort to broaden the kind of information patients can see about the doctors who treat them. About 25 states have adopted similar rules.

What does North Carolina’s health care industry think about the proposal?

[T]he measure has met opposition from doctors and hospitals, the insurers who write their medical malpractice policies and the lawyers who defend them against patient lawsuits…

The hearing Monday drew 32 speakers, with 24 speaking against the board’s plan to post all payments, no matter how large or small, going back seven years.

No doubt some of those settlements involved no wrongdoing by the defendants. But does that mean we should deny patients all such information?

Oh, and another, unrelated story in today’s News & Observer reports:

Eighteen patients who had operations at Duke hospitals in 2004 sued Duke University Health System on Monday, charging that it committed fraud and negligence in connection with the patients’ exposure to surgical instruments mistakenly washed in elevator hydraulic fluid…

Hospital officials cited tests it conducted, which it said showed the instruments were sterile and that microscopic concentrations of fluid that remained on the tools posed no risk to patients.

GAO Issues Report on Privacy

This week, for a hearing in the Senate Homeland Security and Government Reform Committee, the Government Accountability Office released a report on privacy titled “Alternatives Exist for Enhancing Protection of Personally Identifiable Information.” (GAO testimony based on the report is here.) I served on a National Academy of Sciences “Expert Panel” that gave the GAO some perspectives on issues related to the Privacy Act.

The report had three main conclusions, with my comments:

The Privacy Act’s definition of a “system of records” (any grouping of records containing personal information retrieved by individual identifier), which sets the scope of the act’s protections, does not always apply whenever personal information is obtained and processed by federal agencies. One alternative to address this concern would be revising the system-of-records definition to cover all personally identifiable information collected, used, and maintained systematically by the federal government.

The “system of records” definition has indeed fallen out of date. Thanks to the growth of search and other technological developments, records not organized by personal identifier can be accessed and used by the federal government, but they fall outside the purview of the Privacy Act. This should change. The report also highlights the fact that data used by the federal government, but held by information resellers, escapes the purview of the Privacy Act. This should also change.

According to generally accepted privacy principles of purpose specification, collection limitation, and use limitation, the collection of personal information should be limited, and its use should be limited to a specified purpose. Yet, current laws and guidance impose only the modest requirements in these areas… . Alternatives to address this area of concern include requiring agencies to justify the collection and use of key elements of personally identifiable information and to establish agreements before sharing such information with other agencies.

Once they have collected it, federal agencies can do anything they want with personal information simply by declaring their plan to do so in the Federal Register through a “System of Records Notice” or “SORN.” The statements agencies may make when they collect information do not bind them in the slightest. This is wrong and it should change. GAO’s recommendations to limit collection and sharing of information are rather tepid, alas, and they wouldn’t change agencies’ institutional incentives to over-collect and promiscuously share the personal information of the citizenry.

Privacy Act notices may not effectively inform the public about government uses of personal information. For example, system-of-records notices published in the Federal Register (the government’s official vehicle for issuing public notices) may be difficult for the general public to fully understand. Layered notices, which provide only the most important summary facts up front, have been used as a solution in the private sector. In addition, publishing such notices at a central location on the Web would help make them more accessible.

It’s true that Privacy Act notices don’t inform the public well. They are obscurely written documents in an obscure publication. But I’m not sure that the publication of “layered notices” would be an improvement. Sure, there’s a consensus among government types that layered notices are the next big thing, but I don’t believe that they will change citizen understanding or behavior in any significant respect. Notices are also not terribly relevant in the government environment because a person can’t decline to do business with a government based on its privacy practices or promises.

There’s more to learn on “notice” and its importance or relevance for getting people more privacy. The thing we know is that reducing data collection and use leads directly to privacy. Getting policymakers to understand the privacy costs they’re imposing on the public would be as effective, if not more, than notifying the public about what’s been done to them after a policy is made and the horse is out of the barn.