• “Mortgage Amortization and Wealth Accumulation,” by Asaf Bernstein and Peter Koudijs. April 2020. SSRN #3569252.

An important distinction between neoclassical and behavioral economic analysis is the latter’s emphasis on “default effects,” the tendency of people to remain with whatever situation they were originally assigned. The most famous real-world example is the tendency of individuals to save more in employer-sponsored 401k retirement-savings plans if they are enrolled automatically but can opt out relative to when they are not enrolled automatically but can opt in. Neoclassical economists have responded to this by arguing that such nudges could be offset by a decrease in savings outside the retirement framework, negating the benefit of automatic enrollment.

This paper examines a similar situation: consumers who are forced to purchase mortgages that amortize the principal rather than use mortgages that are interest-only with no amortization. In a neoclassical world, mortgage and non-mortgage savings are fungible. If the neoclassicals are right, forced savings through mortgages would be offset by less savings outside the mortgage framework, whereas if the behavioralists are right net savings will increase.

In January 2013 the Netherlands implemented a requirement that new mortgages be fully amortizing. Prior to 2013, most loans were not fully amortizing, leaving homeowners to refinance or make a balloon payment at maturity. This policy change resulted in a substantial increase in monthly mortgage payments. The Dutch maintain careful administrative records on other consumption and savings, so researchers can use those data to determine if increased mortgage savings are being offset by other savings decreases.

Four years after the regulation, the researchers found that the increased mortgage payments were not offset by reduced savings elsewhere. The policy increased net worth overall. From a quarter to a third of the increased savings was financed through higher labor-market earnings and two-thirds to three-quarters was the result of decreased consumption.

The paper goes to great lengths to ask whether these results are the product of factors other than the mortgage policy. Did those who wanted to save less purchase their homes just before the regulation went into effect? There is no evidence of “bunching” in the number of transactions in the months prior to the regulation. The findings hold for households with substantial liquid assets (suggesting the results are not caused by just non-savers) and across all ages. — Peter Van Doren

• “Are Corporate Payouts Abnormally High in the 2000s?” by Kathleen Kahle and René M. Stulz. April 2020. NBER #26958.

• “Why Does Equity Capital Flow Out of High Tobin’s q Industries?” by Dong Lee, Han Shin, and René M. Stulz. February 2020. SSRN#3535841.

Corporate profits and their division between dividends, stock repurchases, and reinvestment are a source of concern for members of Congress from both parties. These two papers look at issues relevant to that concern.

The first paper asks whether payouts (rather than retention of excess cash flow within the firm) are larger now than in the past. The answer appears to be yes. In the 2000s, annual aggregate inflation-adjusted payouts were three times their pre-2000 level and increased as a percentage of assets (2.7% for 1971–1999 versus 4.1% for 2000–2017) and as a percentage of operating income (18.9% for 1971–1999 versus 32.4% for 2000–2017).

The payouts are higher because firms earn more and pay out more of what they earn. Some 38% of the increase in payouts is from higher earnings and 62% from a higher payout rate, which is exclusively from stock repurchase instead of dividends. Dividends average 14.4% of operating income from 1971 to 1999 and 14% from 2000 to 2017. Net stock repurchases averaged 4.8% of operating income before 2000 and 18.3% from 2000 to 2017.

To assess whether the fundamentals that govern corporate financial behavior have changed, the authors estimate a traditional econometric model of payouts with data from 1971–1999 and then use the results to predict current payouts. The model predicts that real aggregate payouts in 2017 should be $784 billion; actual payouts were $734 billion. So, our understanding of current corporate behavior does not require a complete rethinking of conventional financial theory. Higher payouts are the result of changes in the values of variables that historically have explained corporate payouts: increases in firm age, size, and cash holdings, and decreases in leverage. Corporations are investing less as well, but that is less important in explaining the increase in payouts than the four firm variables.

The second paper examines the transformation of the relationship between Tobin’s q (equity market value divided by book value) and capital flows. Capital flowed into industries with higher Tobin’s q over the period 1971–1996. But from 1997 to 2014, capital flowed out of high-q industries. The change is from the repurchase of stock after 1997.

These results make little sense if high-q industries are the ones with the best investment opportunities and competition leads them to expand up to the point where those opportunities are used up. However, these results do make sense if the dominant firms in high-q industries draw rents from scarce assets, so that their high q reflects their ability to collect rents rather than an investment opportunity. As long as their cash flows from rents are high enough and management has incentives to maximize shareholder wealth, it is optimal for these firms to use the cash flows to fund payouts. The funds paid out can then be used by investors to invest where their funds have better uses, which is more efficient than if the firms use these funds to invest in poor projects. — P.V.D.

• “Did Banks Pay ‘Fair’ Return to Taxpayers on TARP?” by Thomas Flanagan and Amiyatosh Purnanandam. May 2020. SSRN #3595763.

The U.S. Treasury pumped hundreds of billions of dollars into the country’s financial firms in 2008–2009 to stabilize the financial system under its Troubled Asset Relief Program (TARP). The conventional wisdom is that the program was fair to taxpayers because it ultimately made money. That is, the $426.35 billion in loans given out during the Great Recession was eclipsed by the $441.7 billion the Treasury received in repayments and from the sale of equities received in exchange for the loans, even after factoring in the $9.5 billion loss on money lent to the auto industry.

The authors of this paper use a different notion of what a fair return should be. They compare the returns realized by the Treasury to what would have been received from market investments with similar uncertainty of repayment over the same period. The preferred equity issued under TARP had the same seniority as the existing preferred equity of the recipient banks, so the market-based returns on the existing preferred equity of the same banks over the same time horizon provide an ideal comparison to TARP’s return.

TARP recipients paid 11% annualized return to taxpayers, while the banks’ preferred equity annualized return was 39% over the same time horizon. In dollar terms, the difference was almost $60 billion per year. Also, bondholders earned an annualized return of 20% per year despite the lower risk. By this measure, taxpayers were shortchanged. — P.V.D.

• “Encouraging Equity Investments in Medium-Sized Businesses,” by Douglas Elliott. Oliver Wyman Policy Paper. May 29, 2020.

In this succinct paper, Douglas Elliott makes the cogent and possibly urgent point that since the global pandemic began, the U.S. economy has transitioned from a corporate liquidity problem — which the Federal Reserve successfully alleviated — to the cusp of a corporate solvency problem. Unfortunately, solving the new problem will be much more difficult and likely cost more money than the liquidity problem.

Virtually no one anticipated the nationwide shutdown from COVID-19. A wide swath of businesses saw demand for their goods crater or were legally obligated to shut down. For instance, passenger air travel declined 95% in a single month and cruise ships shut down entirely. In most states, restaurants were limited to selling only takeout meals, and even that was impeded by curfews. Most brick-and-mortar retailers such as clothing stores also saw demand plummet.

While large corporations were able to tap credit markets to cover their liquidity needs, many mid-sized and small businesses were not able to do so immediately. To alleviate their capital needs, Congress enacted the Coronavirus Aid, Relief, and Economic Security Act, which created (among other things) the Payroll Protection Program, a funding mechanism that provided small and medium-sized businesses forgivable loans to cover payroll and a portion of other costs for eight weeks.

The Federal Reserve also created a variety of credit facilities to help businesses obtain financing, although credit markets ultimately rebounded from the early days of the crisis and most viable businesses were able to obtain financing from private lenders. Some aver that the Fed’s willingness to jump into the market helped the private sector to resume lending.

While the short-term liquidity crunch has passed, we still do not know how long the COVID crisis and the concomitant quarantines will continue, stifling consumer demand across a wide variety of goods and services. The personal savings rate in the United States in May 2020 was 32%, quadruple the level of the previous May and indicating a tremendous decline in consumer demand. Even the most optimistic scenario suggests that vaccines against the SARS-CoV‑2 virus — necessary to return the economy to something approaching normalcy — will not be widely available until well into 2021, if then.

It is also worth noting that not all the decline in consumer spending is the result of deferred spending. While people who had planned to buy a car in the second quarter of 2020 will likely do so at some point in the next year, they will not consume more restaurant meals or haircuts to make up for the lost consumption when the pandemic raged. In those cases, the virus and quarantine effectively destroyed demand. The delay until a durable recovery begets a bigger problem, Elliott observes: numerous businesses will become effectively bankrupt as a result, and those situations will need to be adjudicated in some way.

Should they liquidate? / For businesses that end up in that situation in the next year, we may not want there to be a simple liquidation. Many of these businesses would be viable in an ordinary economy. For instance, a restaurant with a favorable lease in a well-trafficked area in the business center of a community that earned a tidy profit before the pandemic will likely return to profitability post-pandemic. Ideally, we would like the restaurant to be managed post-crisis by the same people who ran it pre-crisis because that combination appeared to work. Its creditors likely would not object. They will all be better off if it reopens in a form closely resembling its previous form rather than a bankruptcy judge ordering it liquidated and waiting for a new entrepreneur to obtain the lease, build his own operation, and open a new restaurant a year or two (or more) later.

Fortunately, the U.S. bankruptcy code has Chapter 11, which allows businesses to reorganize and remain open. If it were a big firm, the creditors would receive equity in the company and existing shareholders would be wiped out. Elliott is more concerned about smaller businesses, where it is more difficult to conceive of any resolution that gives creditors an actual ownership stake in a family business. In this case, the restaurant’s creditors would — ideally — see the money owed to them reduced to some degree and in return the creditors would receive some share of future revenue. In either case, both the debtor and creditor would share in the money lost because of the pandemic, but also in the post-pandemic revenue that is maximized by using the existing resources in their most efficient way. Both are better off as a result.

While that would be the ideal outcome, it may not be practical in many situations. Elliott notes that there could be hundreds of thousands of potentially bankrupt businesses that will need their bankruptcy to be adjudicated. Our federal bankruptcy courts do not have the capacity to deal with such an increase in demand. Even the model of “prepackaged” bankruptcies, where creditors negotiate an outcome and present it to the court to get its official blessing, may not be sufficiently expedient.

One solution, Elliott offers, would be to quickly create some sort of remediation program whereby the courts deputize arbitrators (retired bankruptcy judges or trustees) who help the businesses and debtors reach an agreement. The court would then give the arrangement its imprimatur without its scarce resources being occupied.

Wheat from chaff / Of course, the world will be sufficiently different post-crisis and many businesses will not be salvageable via reorganization. Some businesses will simply be unable to respond to the new environment, especially those that were not doing so well before the pandemic. For instance, it seems unlikely that we will have nearly the demand for cruises post-pandemic, and the least solvent of those businesses will likely go under.

We want to avoid taking steps that would keep nonviable businesses going indefinitely. They would find themselves working to make payments to their debtors and would not have enough money to invest or plan for the future. They would be zombie businesses and keeping them alive would waste investment capital that could go to a better business.

Of course, it can be difficult to readily discern between viable businesses and future zombies. Some restaurants that did fine pre-crisis will not be successful post-crisis, for instance. But any new system we set up to adjudicate these debt issues will not always be able to discern ongoing good businesses from bad businesses. Maybe the cash flow will make it obvious which category a restaurant belongs in, but not always. For this, Elliott suggests dusting off a page from the 1980s–1990s savings-and-loan crisis playbook and instituting some sort of Resolution Trust Authority to make those determinations.

If we do not want a (quasi) government entity picking winners and losers the next time we get in such a mess, Elliott suggests the government might offer insurance against a tail risk economic event. For instance, businesses could buy a type of insurance that would pay a fraction of their losses if gross domestic product were to fall by more than 10% in a quarter. We would then try to ex-ante limit government bailouts in such a scenario.

What is important is to recognize that in this unprecedented economic crisis we have a potential time-inconsistency problem. Government needs to take care that anything it does to salve the short-term economic pain does not concomitantly create long-term problems or establish a precedent that would complicate the country dealing with a similar crisis in the future. This is a message that few want to hear now, but Elliott points out that we may be forced to confront this problem again in the not-so-distant future.

— Ike Brannon