Policy for a Stronger Recovery J. Bradford DeLong Professor of Economics, U.C. Berkeley Research Associate, NBER September 30, 2009
A Little Background About a year and a half ago—in the days after the forced merger of Bear Stearns into J.P. MorganChase, say—there was a near consensus of economists that an additional dose of expansionary fiscal policy was unlikely to be necessary. The Congress had passed a first round of tax cut-based stimulus, the impact of which in the summer of 2008 is clearly visible in disposable personal income and perhaps visible in the tracks of estimated monthly real GDP. The near-consensus belief back then, however, was that that was the only expansionary discretionary fiscal policy move that was appropriate. With the Bear Stearns forced merger it appeared that the Federal Reserve and the Treasury had settled on a policy: they would punish as severely as they could the shareholders of and the managers at institutions too-big-tofail that required rescue, but that they would insulate bondholders and counterparties. The incentives to avoid bankruptcy would thus be concentrated on those who actually had power to do something to manage organizational risk. As for the rest—well, the markets interpreted the forced merger as the Federal Reserve guaranteeing and making riskless essentially all the unsecured debt of all the large commercial and investment banks in the country.
1
Figure 1: (Nominal) Disposable Personal Income
Figure 2: Monthly Real GDP Estimates from Macroeconomic Advisors
2
The resulting “approaching liquidity tsunami,” as more than one senior policymaker described it to me, meant that the risk of a deep recession was very low—or so the situation looked in the spring of 2008.
Late 2008ʼs Need for Expansionary Fiscal Policy By the late summer of 2008 things looked significantly different. The taxbased expansionary fiscal policy of early 2008 had had less than the desired effect—perhaps it had prevented a decline in the economy and kept things marching in place, but it effect was not overwhelming and not entirely obvious. It was clear that the formal announcement that the economy had fallen into recession was only a matter of time. By August 2008 Lawrence Summers was writing of a gap between actual and sustainable production of $300 billion at an annual rate, forecasting that that gap was likely to more than double over the following year, and predicting sustained weakness thereafter—“unemployment peaked nearly two years after the end of the last recession, output and employment are likely to remain below their potential levels for several years in the best of circumstances…” in a time when “the remaining scope for monetary policy to stimulate the US economy is surely very limited…”1 Take an initial output gap growing from $300 to $600 billion over the first year and then declining to zero over the next three and you have a cumulative output gap of $1,350 billion in a situation in which monetary policy on it own can do little to correct it. Suppose that a prudent use of fiscal policy would be to enlarge the government’s budget deficits by a third of the forecast output gap, and you have an estimate of the appropriate size of expansionary fiscal policy as the situation looked in August 2008: $450 billion in cumulative deficit spending spread out over the next four years. Then came the nationalization of Fannie Mae and Freddie Mac on September 7, 2008; the bankruptcy of Lehman Brothers on September 15, 2008; and the nationalization of AIG on September 22, 2008. In the 1
Lawrence Summers (2008), “The Big Freeze, Part IV: A U.S. Recovery,” Financial Times (August 6, 2008) http://tinyurl.com/dl20090927a.
3
aftermath it was immediately clear that the recession problem was at least twice as bad as it had looked in August, and over the next four and a half months until the February 17, 2009 signing of the ARRA the magnitude of the likely cumulative output gap doubled again as the magnitude of the financial crisis’s impact on the real economy became clear. If $450 billion was the appropriate size of a short-term deficit-spending program for the $1,350 billion cumulative output gap that Lawrence Summers had anticipated as of August 2008, then the appropriate size of the boost to short-term deficit spending as of February 2009 was $1.8 trillion (over three to four years). What we got was a cumulative number of $600 billion—roughly 1/3 aid to states, 1/3 tax cuts (in a good-faith effort by the Obama administration to propose a bipartisan plan that legislators of both parties could sign on to), and 1/3 infrastructure and other direct government purchases intended not so much to slow the decline as rather to boost the recovery. We also got an extension of the AMT and other measures that no economist I have talked to believes are properly counted as part of an effective fiscal boost under any currently-live theory of how the economy works. Figure an increase in deficits of $200 billion per year spread out over the next three years. At the technocratic level, the disproportion between the size of the response and the magnitude of the need is obvious.
Todayʼs Arguments Against More Fiscal Expansion Now if you go a little bit south to Lafayette Park and, addressing the air, ask why it is that we did not pass a larger short-term deficit-spending fiscal boost program of $1.2 or $1.8 trillion last January and February and why we are not acting to boost it now, I hear four answers back on the wind: 1. This was the most that we could get sixty senators to vote for—and with a Senate that, in Majority Leader Harry Reid’s words, takes forty-eight hours to flush the toilet we need to spend our time on legislative initiatives that might pass, rather than on those that certainly will not.
4
2. Further expansionary fiscal policy would be counterproductive in this current situation because of the long-run U.S. budgetary and global balance-of-payments imbalances. More short-run deficit spending would require the U.S. to issue more debt which would cause a sharp spike in U.S. long-term interest and a flight from the dollar that would generate a much bigger crisis and deeper depression—as Austria’s issuance of huge amounts of additional debt in 1931 set off the wave of crises that turned the recession of 1929-1931 in Europe into the European half of the Great Depression. 3. Further expansionary fiscal policy would be counterproductive because it never works—because it is theoretically impossible for it to work—because it is a “fairy tale.” 4. The ARRA is only one of a large number of initiatives to stimulate the economy outside of the normal open-market operations monetary expansion framework. When you include the likely effects of all of the acronyms—TARP, PPIP, MMIFL, TALF, CPLF, TAC, etc.—you find that even though we have only $600 billion of cumulative expansionary fiscal policy we have much more in terms of total non-standard-monetary stabilization policy. Argument #1 I will pass over. I understand why Rhode Island and Delaware have as many senators as New York or Florida or Texas or California: it seemed a reasonable price to pay back in 1787 to keep not just Montreal but also Providence and Wilmington from becoming British imperial fortresses and bases on the North American continent. I understand that a “cooling” chamber might wish to have an effective supermajority requirement. I don’t understand why in a good system it takes the votes of 60% of senators rather than of senators representing 60% of the people to actually get an up-or-down vote. But that is not my area.
5
Argument #2 is, I think, a genuine thing to fear. But if that fear were to cease being a nameless dread and instead take a shape and a name, the first sign would be an unwillingness on the part of global investors to hold U.S. Treasury debt. But right now the U.S. government can borrow for ten years at 3.34% in nominal dollars—much less than the projected growth rate of nominal GDP over the next decade at between 5% and 6% a year. The time since the summer of 2007 has seen a collapse in the value of private securities and a substantial elevation in the price of U.S. government securities. The first rule of the market is that the market’s prices are there to signal what things are more valuable and that we should make more of. Right now the market is sending us a very strong signal that it really wants us to make more U.S. Treasury debt—to undertake more short-term deficit spending. And when argument 2 ceases being a nameless dread but takes form and shape in an upward trend in interest rates on long-term U.S. Treasuries, I will be the first to say that the global bond market is cutting off our running room for more expansionary fiscal policy. Right now, however, it is not. Argument #3—that there is some deep principle in economic theory somewhere that says something that prevents debt-financed government spending from boosting employment and output—is an argument that, by this stage, I can do nothing but laugh at. When the venture capitalists of Silicon Valley decided to give their money to high-tech engineers in the 1990s so they could spend it trying to figure out how to make money off of the internet, employment and production rose (even though few of them figured out how to do so). When the financial engineers of Countrywide in the 2000s decided to give investors’ money to construction companies to build more houses, employment and production rose (even though the promises to investors of healthy returns with little risk were wrong). Milton Friedman’s quantity-theory-of-money monetarism says that the reason open-market operations that expand the money stock boost spending and employment is because once people have more money in their pockets they step up their spending. Boosts to employment and production come when any group with substantial money in an economy decides to boost its rate of spending—and, at this level, the government’s money is as good as anybody else’s.
6
And when I try to read the arguments of those making Argument #3, my head explodes. The kindest thing I can say is that they must not have spent even half an hour thinking about the issues. John Cochrane of the University of Chicago writes that models in which fiscal policy affects anything are logically inconsistent because they require that people’s plans of how much to spend exceed what their incomes actually turn out to be. But Chicago models of an earlier generation like Milton Friedman’s are full of situations in which planned expenditure is greater than or less than income as people try to run down or build up their cash balances. Eugene Fama of the University of Chicago writes that any increase in government purchases must be automatically offset by an equal decrease in business investment or household consumption spending—a conclusion that had eluded every other monetary economist since the 1910s, and a conclusion which would make not just fiscal policy but monetary policy impotent.2 At this stage all I can do is two things. First, I can gesture at the Republican office holders and policy advisors of last year—at the Doug Holtz-Eakins, the Mark Zandis, the Phil Swagels and the others who are now saying that you should be “skeptical” of “anyone who tells you [expansionary fiscal policy] has had no impact.” Second, I can point out that had John McCain won the 2008 presidential election the Republican administration would have no more hesitation in proposing expansionary fiscal policy than they did in 2008, 2003, and 2001, and that the only reason the views on macroeconomics of Cochrane, Fama and company are now getting a hearing as Republican witnesses in congressional hearings is 2
Note that neither Cochrane nor Fama are the right-wing fringe of the economics profession right now. The right wing fringe is composed of the supporters of Ed Prescott, Ed Prescott who used to teach at the University of Chicago before moving to Arizona State. He says that Fama’s conclusion that monetary policy does not affect output is in fact correct: in his view the Great Depression was not caused, as Milton Friedman thought, by the bank failure-induced collapse of the money stock but by Herbert Hoover’s “anti-market, anti-globalization, anti-immigration, pro- cartelization policies were instituted… [which] created a great depression.” Chicago economists of an earlier generation, like Jacob Viner, had no doubt at all in the correctness of their policy advice: that the Great Depression had been caused by unbalanced monetary deflation and needed to be cured by expansionary monetary and fiscal policy.
7
that Republican legislators need some form of ideological cover for their just-say-no-to-everything-whether-it-is-good-for-the-country-or-not legislative strategy. Argument #4—that the asset purchase, asset guarantee, and bank recapitalization policies are about to have a big effect and boost the recovery—is one that I really wish were true, but I just don’t see evidence of it, at least not right now. Yes, spreads have narrowed. But asset values are still low: the S&P 500 stands about where it did early last October, after Lehman, after AIG. The banking-sector policies were supposed to boost the confidence and the risk tolerance of the private market in order to get the engine of private sector lending and borrowing and spending rolling again. Things are much better than they were at the start of March 2009, it is true. But as measured by the mark of asset prices, they are no better than they were in October 2008. And it is by asset prices that the banking-sector support policies should be judged. The right way to look at monetary and financial policy is that it has, ever since 1825, been focused on manipulating asset prices: the central bank buys and sells and guarantees and regulates and subsidizes and nationalizes with an eye toward pushing the prices of financial assets to levels where businesses seeking to raise capital to build capacity and households seeking to spend out of wealth together can issue new assets and so access enough money to push their spending to a level that gets the economy to full employment, or at least out of depression. The policies are always sold as opaque technocratic adjustments to the “money stock” or to a “federal funds interest rate” that real people do not see and is of concern only to bankers. But the policies are and always have been truly aimed at manipulating asset prices. We may believe in a market economy. But since 1825 we have also believed that asset prices are too important to be left to the market to determine when their free market levels produced either depression unemployment or runaway inflation. Thus the thing to focus on is that the prices of risky financial assets are very low—not as low as they were last March, when the S&P 500 kissed a level of 667, but still very low. Why are they so low? The answer is that
8
the risk tolerance of the private market has collapsed. For example, consider what the University of Chicago’s Nobel Prize-winning economist Bob Lucas told Tom Keene of Bloomberg last March 30—that he was 100% in cash: LUCAS: [T]here is no question that fear is what this liquidity crisis is. I mean the reason I got into money [with my portfolio] is that I got afraid to leave my pension fund in other securities. So I’m sitting there with a portfolio full of zero-yield stuff just because I’m afraid to do anything else. I think there are millions of people like me. KEENE: What will be the signal for Robert Lucas to go back into the markets...? LUCAS: I don’t know. Robert Rubin made a joke about that in the first session today. Nobody knows...
Now let’s pick on Lucas because he is not here to defend himself, Earlier in the interview, he had told Tom Keene: LUCAS: Our economy’s got a remarkable ability to return to its long term growth trend. And for most of the depressions we’ve had or recessions, the return has been quick. Two or three, four years...
Lucas says that it is highly likely that the U.S. economy will be back to normal in three or four years, with a normal level of unemployment, a normal share of profits in national income—and a normal level of dividends and capital gains. This presumably means that stock prices will also be back to a normal multiple of long-run earnings, which means a year-2015 S&P of 2000 or so, compared to its current value of 1044 or its 2009 low of 667. Investing in the S&P 500 for a four-year horizon now is risky, certainly, but the expected return is high: 15% per year, if you believe Lucas’s forecast. And holding your money in cash is not all that safe either: the scenarios I can envision in which the S&P 500 is at a real value in 2015 corresponding to the value that 667 buys today are scenarios in which inflation has eaten away most of the value of cash.
9
So what is Lucas doing holding his portfolio in cash? Has risk suddenly increased to an extraordinary extent to force the equity share of his portfolio down from 70% to 0% in spite of the huge jump in expected sixyear returns? Has his personal tolerance for risk suddenly collapsed? No. He is irrationally panicked. And, as he says, there are millions like him. Until they recover from their panic, even a perfectly constructed banking and financial system will not produce the asset prices needed for private investment spending to drive us to full employment.3 Thus the banking-sector support policies have not been a bust—they have surely kept things from getting much worse. But they have not done much if anything that promises to close the output gap.
What Should We Do Now? The argument that more expansionary fiscal policies should not be tried because it is theoretically impossible for them to work fails. The argument that more expansionary fiscal policies should not be tried because the unstable nature of global imbalances and U.S. long-run fiscal deficits has us teetering on the edge of a Credit Anstalt-like currency crisis disaster
3
An aside. Lucas’s irrational panic—and his observation that there are millions of investors like him—makes the part of his interview with Tom Keene in which he attacks Robert Shiller and George Akerlof and their book Animal Spirits quite puzzling. Here is what Lucas says: “I just don’t get it. I mean look at the Black_Scholes formula. People come up with a formula for pricing options, just out of purely mathematical reasoning, and then it turns out it fits certain data amazingly well. And it’s been incredibly useful to people. Now it’s not useful for everything, but for what it does it’s a huge advance in human knowledge. Now what’s the behavioral finance contribution? They reinforce the idea of skepticism. Well, skepticism, it’s easy to be a skeptic. What’s harder is to tell somebody how to do something they didn’t know how to do yesterday. That’s what Black and Scholes did...” Lucas seems to miss the entire big point. Black-Scholes tells you how to do things only if the trades of the non-von Neumann-Morgenstern agents in the economy—that's him—cancel each other out. If they don't cancel each other out—if there are, as Lucas says, “millions of people like me,” then a whole bunch of banks running off of Black-Scholes and similar models create a lot of systemic risk, and then 10% unemployment. That’s the problem Akerlof and Shiller are trying to grapple with.
10
fails. The argument that banking policies have been successful enough that we do not need more expansionary fiscal policy fails.
Figure 3: Troika Forecast of the Unemployment Rate as of August 2009
The Congressional Budget Office currently forecasts that the unemployment rate will average 10.2% in 2010, 9.1% in 2011, and 7.7% in 2012 before returning to its normal range with an average of 5.1% in 2013. The administration’s Troika forecast is very similar. Things will almost surely be either significantly better or significantly worse than the forecast—it is a forecast, after all—but the right thing to do is to plan as if the forecast will come true, and then adjust later. If you are happy with that forecast and think that it is appropriate for the U.S. economy over the next four years, then no further support for the
11
recovery appears needed at this time. If you are unhappy with that forecast, then additional federal government action is definitely advisable.
Figure 4: Past and Projected Employment-Population Ratio
What kind of action, however? The obvious would be additional short-term deficit spending on the federal level: (1) Triggers to extend the existing expansionary fiscal policy measures should the unemployment rate not decline rapidly: since we remember the history of the 1937-1938 episode, we are hopefully not condemned to repeat it. (2) An expansion not in length but in flow of the current fiscal boost package: last January a number of us were saying that an $800 billion
12
cumulative fiscal boost was OK—but that there also ought to be a trigger in the budget resolution so that if unemployment rose near 10% the fall reconciliation bill could be used to top off the program. That didn’t happen. It ought to have happened. It would be nice to make it happen—and there is a deal to be struck with more deficit spending in the short-term and tax-increase or spending-cut triggers in the long term should the deficit not return to sustainable levels after the recession passes. (3) Less obvious would be measures to aid useful deficit spending in other levels of government. During this recession the states have, as Paul Krugman puts it, turned into fifty little (and not so little) Herbert Hoovers. The obvious policy to enable states that want to avoid counterproductive budget-cutting in this recession to do so is: •
Have the federal government support the prices of deficit-spending bonds issued by state governments that also put credible and automatic amortization plans in place.
This support could be provided either at the level of the Treasury—with the Treasury Department approving state fiscal policies as sustainable in the long term and thus qualifying for loan guarantees—or at the level of the Federal Reserve—with the Federal Reserve offering to support the prices of state deficit-spending bonds that come attached to legislated sustainable state-level fiscal policies. (4) Also a possibility: bringing forward long-term investments that we ought to be making over the next generation. The people I talk to at Berkeley who actually know what they are talking about say that over the past decade and a half since the Senate rejected Al Gore’s BTU tax the 3 degree Fahrenheit warmer world in 2150 has probably slipped out of our grasp, and that the good possibility going forward is a 7 degree Fahrenheit warmer world in 2150. To get there the U.S. has to lead—take the first steps—and then work hard to pull the rest of the world along to an appropriate global warming policy.
13
The best way to get there would be a carbon tax. A somewhat worse way would be a cap-and-trade system that grandfathers in many current highly inefficient uses of open carbon cycle energy. A still worse way would be for the EPA to start regulating greenhouse gases as pollutants. We would like to use the effective carrot of the market rather than the stick of command-and-control regulation, but the Senate may keep us from doing so. In that case—if we aren’t providing the incentives for businesses and people to make the investments in closed-carbon cycle and other green energy technologies through a carbon tax or a cap-and-trade system—then the government will need to make those investments or provide separate incentive programs to induce people to make them. How big? $5,000 per household in commercial, industrial, and residential cost-effective energy investments does not seem out of line. (We’ve done $20,000 over the past five years in our house and are getting 6% per year at current PG&E electricity prices.) And that would be $500 billion nationwide. 4,066 words: September 30, 2009
14