Banking Regulation

• “The Gramm‐​Leach‐​Bliley Act of 1999: A Bridge Too Far? Or Not Far Enough?” by Lawrence J. White. May 2011. SSRN #1836668.
• “Robust Capital Regulation,” by Viral Acharya, Hamid Mehran, Til Schuermann, and Anjan Thakor. April 2011. SSRN #1822333.

Many commentators, including former Federal Reserve Board chairman Paul Volcker, have blamed the recent financial crisis and subsequent recession on the 1999 Gramm‐​Leach‐​Bliley Act (GLB), which eliminated the legal barriers between investment (securities) and commercial (traditional deposits and loans) banking and insurance. In a recent paper, New York University economist Lawrence White argues that this blame is misplaced for two reasons.

First, the Depression‐​era theory that investment and commercial banking should be separated has long been discredited by academic economists. The theory alleges that the stock market crash of 1929 was the result of investment banks selling poor‐​quality stocks to an uninformed public in order to help the stock‐​issuing companies repay loans from the investment banks’ affiliated commercial banks. Randall Kroszner and Raghuram Rajan (“Is the Glass‐​Steagall Act Justified,” American Economic Review, Vol. 84, No. 4) debunked this theory nearly two decades ago when they showed that securities underwritten by banks with affiliates were no more likely to default than similar securities issued by independent investment banks during 1924–1940.

Second, GLB simply granted congressional recognition of regulatory decisions in the 1970s and 1980s that blurred the distinctions between investment and commercial banking. Those decisions had been incorporated into the marketplace long before 1999, as evidenced by the creation of money market funds, the securitization of mortgages, and the direct access of corporations to loans through the commercial paper market.

For White, the financial crisis was the result of high‐​leveraged lending to the housing sector, and those loans ultimately went sour with the 2006–2007 popping of the real estate bubble. What’s relevant to the GLB discussion is that those losses did not occur because commercial banks underwrote corporate securities, or traded for their own account, or operated hedge funds. The losses occurred because both commercial and investment banks invested in mortgage‐​related bonds, which was legal (and encouraged) long before GLB. Hence, White concludes, the reimposition of Glass‐​Steagall Act rules separating investment banking from commercial banking would “not even be a case of locking the barn door after the horses have run out. Instead, it would be a case of closing a set of side doors that the horses hardly notice.”

This is not to say that GLB doesn’t deserve some criticism relevant to the financial crisis. GLB increased the barriers between commerce and banking, which hurt low‐​income consumers. In the late 1990s, Wal‐​Mart wanted to enter retail banking, both in an effort to lower its credit card costs and because it saw a profit opportunity. Existing banks, existing retailers, labor unions, and other opponents of Wal‐​Mart succeeded in inserting language into GLB that prevented any acquisition of a unitary thrift holding company by a commercial or industrial company, thereby blocking Wal‐​Mart’s plans. Low‐​income consumers would be better served by aggressive entry by companies like Wal‐​Mart with a proven track record of serving low‐​income customers—indeed, they’d benefit far more from this than programs like the Community Reinvestment Act that “pressure” existing banks to lend to the poor.

Another recent paper, by Viral Acharya et al., examines the role that high‐​leveraged lending played in the financial crisis. It is now consensus that high‐​leverage financing—that is, extensive borrowing by banks, which then used the borrowed money to finance housing and other investments—fueled the housing bubble. Many policy analysts reason that decreased leverage—that is, increased use by banks of their shareholders’ money to finance such investing, instead of borrowing—would allow the banks to survive large asset losses and decrease the need for future government bailouts.

The authors of this paper worry that such a change could ultimately lead to more financial problems. Snakebit uninsured large depositors and debt‐​holders now have powerful incentive to monitor the investment behavior of financial institutions that want to borrow (or have already borrowed) their money; a switch to greater equity financing would dampen that vigilance.

Acharya et al. propose that financial firms be required to have a “mandatory equity buffer”—in effect, a firm “rainy day fund”—funded out of retained earnings. This would insure savings during good times and use of the savings during bad. The equity buffer would be transferred to the normal capital account of the financial institution automatically when prearranged capital levels are breached because of loan losses. If the bank becomes insolvent, the equity buffer would revert to the Federal Deposit Insurance Corporation so that uninsured debt‐​holders would not benefit directly from the extra equity and thus still have incentive to monitor the bank’s loans.

The objection to increased capital requirements is that they would reduce the value of banks. Acharya et al. respond by invoking the insights from a paper by Anat Admati that I described in an earlier Workings Papers column (Winter 2010–2011). Equity is “expensive” only because banks are so highly leveraged and thus risky. More equity reduces risk and thus reduces the “cost” of the equity. The result is that more equity does not reduce the market value of banks.

Housing Markets

• “Did Local Lenders Forecast the Bust? Evidence from the Real Estate Market,” by Kristle Romero Cortes. November 2011. SSRN #1967179.
• “Housing Price Variation and the Convenience Yield to Owning a Home,” by Jason Thomas and Robert Savickas. January 2012. SSRN #1986464.
• “How High Gas Prices Triggered the Housing Crisis: Theory and Empirical Evidence,” by Steven Sexton, JunJie Wu, and David Zilberman. February 2012. University of California Center for Energy and Environmental Economics WP-034.

In “Would a Stricter Fed Policy and Financial Regulation Have Averted the Financial Crisis?” (Cato Policy Analysis 648, October 2009), Jagadeesh Gokhale and I argued that the claims by analysts that they saw the housing bubble and its implications in real time are, for the most part, unsubstantiated by the papers and analyses those analysts issued during the boom. More likely, the claims are the product of “foresight” that comes from looking in the rearview mirror.

However, an interesting paper by Boston College doctoral candidate Kristle Cortes suggests that some analysts may have accurately foreseen the collapse. She argues that local lenders, defined as financial institutions having a branch location in the county in which a mortgage was granted, behaved as if they did see the bubble in real time and reduced their lending as a result.

From 2002 to 2006, a 1–standard deviation increase in housing prices in a ZIP code is associated with a 15 percent decrease in local lending (defined as the share of mortgages coming from lenders with a branch in the county in which a loan is made). In that time period, local lending declined most in areas that had the greatest subsequent decline in housing prices in 2006–2009. Local lending decline is also associated with a decline in the share of local loans held in portfolio. That is, local lenders sold off rather than retained more mortgages in those areas that later experienced the greatest 2006–2009 housing price decline. Local financial institutions behaved as if they perceived housing prices to be excessive and reduced their exposure.

While local banks may have seen the housing bubble for what it was, homebuyers did not. Jason Thomas and Robert Savickas try to discern why the buyers kept buying. One offered explanation is that buyers were tempted by loans they should have never received. Much of the popular discussion of the housing crisis has focused on so‐​called “subprime” loans to people with poor credit histories, but three‐​quarters of the post‐​bubble defaults experienced by Fannie Mae and Freddie Mac have been on higher‐​quality Alt‑A and interest‐​only loans made to borrowers with high credit scores and reasonable loan‐​to‐​value ratios. Policy discussions on avoiding more of these defaults in the future have centered on the features of the lending products: the lack of income and other documentation in the case of Alt‑A and the lack of principal reduction in the case of interest‐​only loans.

Thomas and Savickas point out that these loans were used only in areas of the country with high and/​or rapidly increasing housing values. If the housing boom had not been a bubble with a subsequent bust, the defaults would not have occurred because the houses’ value would have insured that creditors recouped their money. Hence, the authors say, a policy response should not focus on the characteristics of mortgage products, but on the rise and decline in house values.

Unlike other explanations of house value increases that emphasize supply constraints both natural and regulatory, the authors compare the rental and ownership price of housing services. Unlike house prices, rental costs did not soar during the housing boom. So why did large numbers of people opt to buy rather than rent comparable housing? For example, between 2001 and 2006, the real rental cost of equivalent housing in Miami declined and the relative cost of owning rather than renting doubled. Why did people keep buying?

The authors use the term “convenience yield” to describe the difference in the flow of services from owning rather than renting a particular home. Convenience yield consists of two components: a long‐​run component that captures the ability to design the residence the way you want, and a more transitory component: the ability to borrow against home equity at rates that are much lower than uncollateralized loans available to renters. At the bubble’s peak in the mid 2000s, U.S. home equity borrowing was $200 billion per quarter, or 12 percent of U.S. consumption. The ability of homeowners to have higher wealth and consumption than renters reinforced the already prevalent American belief that owning a home is the surest path to wealth creation. This caused more households to buy homes, which increased home prices, borrowing‐​against‐​equity, and convenience yield. But once home prices stopped rising, endogenous convenience yields plummeted so that they now reflect only the much lower long‐​run usage benefits of homeownership. Existing renters no longer use exotic mortgages to become owners and recent buyers have unhappily discovered that they overpaid for ownership because equivalent housing services were available in the rental market at lower prices.

Another question from the housing crash is why the bubble popped when it did. In the above‐​referenced Cato Policy Analysis, Gokhale and I suggested that the 2006 gasoline price spike triggered the collapse. Steven Sexton, JunJie Wu, and David Zilberman present a much more rigorous argument for this idea.

In metropolitan areas where the job market was strong and housing demand was high in the mid‐​2000s, the fewest constraints on new construction—both regulatory and natural—were found at the areas’ far edges and beyond (i.e., the exurbs). Thus, potential homebuyers often engaged in a complex calculus when buying: how much of their paycheck from their city job were they willing to devote to commuting from their affordable suburban or exurban home? The doubling of oil prices from January 2005 to January 2008 unexpectedly raised that cost, hurting homeowners and reducing the value of exurban homes. A 10 percent increase in gasoline prices resulted in a 10 percent decline in suburban relative to urban construction over four years.

In California, the largest median house price declines occurred in jurisdictions furthest from major cities. The 15 jurisdictions that experienced the least house price declines were richer and had gasoline expenditures that were 31 percent lower than those jurisdictions with the largest house price reductions. An analysis of Zillow Home Value Index data for 269 metropolitan areas in the United States concludes that the loss in home value from the 2006 peak to subsequent trough is positively correlated with distance from the central business district.

Wireless Communication

• “Like Deck Chairs on the Titanic: Why Spectrum Reallocation Won’t Avert the Coming Data Crunch But Technology Might Keep the Wireless Industry Afloat,” by Brian J. Love, David J. Love, and James V. Krogmeier. August 2011. SSRN #1914058.

Since 2008, global mobile data traffic has increased an average of 140 percent per year and is expected to grow 26‐​fold by 2015. AT&T alone had a 30‐​fold increase between the third quarter of 2009 and the third quarter of 2010. The reason for this explosion is that smartphones (24 times), tablet computers (122 times), and mobile broadband–equipped laptops (515 times) use much more bandwidth than a simple cell phone.

Economists often argue for markets in spectrum reallocation so that spectrum currently used for radio or television can be sold for more valuable mobile communication use (for example, see Jerry Ellig, “Costs and Consequences,” Fall 2005). But even if the most optimistic reallocation occurs, mobile services bandwidth will increase only three‐​fold, while use has increased 2.5‑fold per year over the last three years. A three‐​fold one‐​time increase will not do the job.

In this paper, the authors argue that technological innovation could enable cell phone companies to move a lot more data across their current spectrum resources, but those companies simply are not that innovative. Their focus is on lobbying the government for more hertz of bandwidth rather than on increasing the number of bits transmitted per hertz. The authors argue that the incentives for technological change rather than simple spectrum allocation should be of concern to policymakers.

The External Costs of Vehicle Weight

• “Pounds that Kill: The External Costs of Vehicle Weight,” by Michael Anderson and Maximilian Auffhammer. June 2011. NBER #17170.

In the early 2000s, there was considerable public attention given to whether the dramatic increase in the use of light trucks was a problem because of their weight and size. As part of the discussion, Regulation published an article that argued that increased sport‐​utility vehicle use saved lives on net because SUVs and other light trucks protect their passengers well (Douglas Coate and James VanderHoff, “The Truth about Light Trucks,” Spring 2001).

The issue no longer garners much attention, but scholars continue to investigate it. In this paper, Michael Anderson and Maximilian Auffhammer argue that each 1,000-pound increase in the weight of a striking vehicle in an accident increases the fatality probability in struck vehicles by 0.09 percent. As the average probability of fatality in the authors’ sample is 0.19 percent, each additional 1,000 lbs. of vehicle weight results in a 46 percent increase in risk. The average car on the road in 2008 was 530 pounds heavier than in 1988. From a Pigouvian perspective, a gas tax of 27 cents per gallon would account for the $35 billion in external costs from the average vehicle weight gain since 1988.

How would such a gas tax compare with the current corporate average fuel economy (CAFE) regime? CAFE imposes a “tax” on the price of a vehicle through higher capital costs while an explicit gas tax collects revenue over time. CAFE increases the price of pickup trucks by 0.6 percent (about $200) while an appropriate Pigouvian safety gas tax would be $4,000 over the life of the vehicle. Light truck frames impose significant risks on other motorists and provide little or no safety benefit for their own occupants according to Anderson and Auffhammer’s statistical analysis. Light truck purchases should be discouraged but the new CAFE regulations that establish different mileage standards for different‐​size vehicles encourage them.