Have We Misdiagnosed the Crisis?
Friday, October 9, 2009
What if the conventional wisdom is wrong? Some basic insights from modern macroeconomics suggest a very different interpretation of recent events.
Most people have a general idea of the events that led to the current recession. Lax lending standards contributed to a housing bubble that peaked in mid-2006. As housing prices began declining, more and more mortgages were defaulted on, and bank balance sheets deteriorated rapidly. In September 2008, Lehman Brothers failed, and the financial crisis seemed to become much worse. A mild recession turned into one of the most severe since the 1930s. But what if the conventional wisdom is wrong? Is it possible that we have fundamentally misdiagnosed the crisis? You might be surprised to learn that some basic insights from modern macroeconomics suggest a very different interpretation of recent events.
I will start with my own somewhat unconventional take on recent events, and then discuss why other economists have reached different conclusions. Keep in mind that this is not merely an exercise in Monday morning quarterbacking; unless we correctly diagnose the problem, we will not be able to come up with effective remedies.
It will be helpful to begin with the concept of nominal GDP, the current dollar value of economic output. Over the past few decades NGDP has grown at just over 5 percent per year. Because real GDP growth averages about 3 percent per year, inflation has recently averaged just over 2 percent. Modern macro theory suggests that monetary policy determines the growth path of nominal GDP. In the long run, monetary policy affects only nominal variables such as inflation and NGDP growth, but in the short run a sudden change in nominal GDP growth can have important real effects, as we are seeing today.
Yes, many foolish subprime loans were made in recent years, but the fall in nominal GDP has turned a $1 trillion fiasco into a $3 to $4 trillion tragedy. It didn’t have to be this bad.
After the magnitude of subprime losses became apparent in the late summer of 2007, NGDP growth slowed, averaging 3 percent during the following nine months. Not surprisingly, real growth slowed as well. When you consider that we also faced a very severe “supply shock” from high oil prices, it is surprising the recession wasn’t more severe during early 2008. The real economy grew slightly during the first half of 2008, despite both the banking crisis and the record oil prices. Then after July 2008 (and before Lehman failed in mid-September) the economy slowed sharply. The recession had been initially concentrated in the subprime real estate markets, but now spread to many inland housing markets.
More ominously, after July 2008 there were many other indicators that money was far too tight. About this time the dollar began rising sharply against the euro; and commodity prices began a sharp decline. Real interest rates rose from less than 1 percent in July to more than 4 percent by November. In the first ten days of October the stock market crashed 23 percent, signaling much more bearish expectations for the economy. The recession spread beyond housing, depressing manufacturing and other sectors. Both real and nominal GDP fell at annual rates of 5 percent to 6 percent in the fourth quarter of 2008 and the first quarter of 2009. By the second quarter of 2009, nominal GDP was 2.4 percent below its level a year earlier, and an astounding 7.5 percent below the long-term trend. Nominal GDP growth is very important during a debt crisis, as loan defaults soar when nominal incomes decline. During 2009 we may well experience the largest decline in nominal income since 1938. Yes, many foolish subprime loans were made in recent years, but the fall in nominal GDP has turned a $1 trillion fiasco into a $3 to $4 trillion tragedy. It didn’t have to be this bad.
Over the past several months, I have been arguing that the general public and even most economists missed a fundamental change in the nature of the crisis during the late summer of 2008. Whereas the initial downturn was mostly caused by “real shocks,” such as banking turmoil and high energy prices, the far more severe second stage of the recession was triggered by a sharp decline in NGDP, which represents a failure of monetary policy. A common cold had turned into pneumonia, but was still being treated like a viral infection. The most controversial part of my thesis, and the hardest to understand, is my argument that monetary policy was actually far too contractionary during late 2008, relative to the needs of the economy. At the time, most economists simply assumed that policy was expansionary because the Fed had cut rates to relatively low levels, and the “monetary base” (which is the money produced by the Fed) rose very dramatically in the fourth quarter of 2008. Neither nominal interest rates nor the money supply, however, is a reliable indicator of the stance of monetary policy.
You might be surprised to learn that economic historians no longer believe that financial instability was the cause of the Great Depression.
To better understand this issue, let’s go back and look at what happened in the Great Depression. After the stock market crashed in late 1929, the Fed gradually cut interest rates, from about 6.5 percent to 1.5 percent. In addition, they sharply raised the monetary base over the following three and one half years. Does this sound familiar? At the time, most people thought that monetary policy was highly expansionary, and that the Depression was caused by financial instability (the stock market crash, banking panics, etc.). Today, as in the 1930s, almost everyone considers the financial crisis to be the proximate cause of the severe contraction that occurred late last year (although experts differ as to the deeper root causes); indeed it is difficult to find anyone promoting alternative explanations. Therefore, you might be surprised to learn that economic historians no longer believe that financial instability was the cause of the Great Depression. Instead, the most accepted explanation, popularized by Milton Friedman and Anna Schwartz, is that the Depression represented a failure of monetary policy.
Friedman and Schwartz argued that both nominal interest rates and the monetary base were unreliable indicators of monetary policy during the Depression. Interest rates fell because of deflation and a weak economy, not because money was easy. In fact, tight money can actually cause interest rates to fall, and did so in December 2007, when a contractionary policy surprise from the Fed (a smaller than expected rate cut) reduced Treasury bond yields on maturities from three months to 30 years. Most people are only familiar with the so-called liquidity effect, the tendency for nominal interest rates to fall when money is added to the economy. But monetary policy also affects real growth and inflation, and in doing so can cause interest rates to move in a “perverse” direction. That happened in the early 1930s, and again in 2007 to 2008.
But what about the money supply? Didn’t that fall in the Great Depression? Actually the quantity of money directly produced by the Fed rose sharply, as the Fed partially accommodated the extra demand for liquidity by the public and banks. Today, we again see a big increase in the monetary base (comprised of bank reserves and cash.) Friedman and Schwartz focused on the so-called monetary aggregates, such as M1 and M2, which are mostly comprised of bank deposits. It is true that M1 and M2 have recently risen, whereas these aggregates fell sharply in the early 1930s. But that doesn’t mean money is easy today, rather the earlier pattern reflected the lack of deposit insurance. Today, FDIC-insured bank deposits are a very safe and liquid refuge in a time of financial instability. Furthermore, in the 1980s most economists concluded that monetary aggregates such as M1 and M2 were not reliable monetary policy indicators.
If neither the money supply nor nominal interest rates are reliable policy indicators, which indicators are reliable?
Of course, not everyone accepts Friedman and Schwartz’s interpretation of the Great Depression. But Federal Reserve Chairman Ben Bernanke does, at least he accepts their argument that the Fed was to blame for the Great Contraction of 1929 to 1933. So I don’t think we can simply say that Fed policy was “obviously” expansionary in late 2008, simply because nominal interest rates were cut and the monetary base increased sharply. That may be the case, but we need to probe more deeply before reaching any conclusions. But if neither the money supply nor nominal interest rates are reliable policy indicators, which indicators are reliable?
For almost a quarter-century, I have been advocating a “forward-looking monetary policy,” where policy makers would target the forecast. Thus, if the Fed’s policy goal were 2 percent inflation, they would adjust monetary policy until the expected rate of inflation was about 2 percent. If inflation was expected to be more than 2 percent, the current policy stance would be too expansionary, and vice versa. More recently, prominent economists such as Lars Svensson of Princeton have also advocated targeting the forecast. Bernanke has not explicitly endorsed this policy, but in a number of statements he has hinted that he looks at things this way. For instance, he recently stated that “the longer run projections of inflation may be interpreted, in turn, as the rate of inflation that FOMC [Federal Open Market Committee] participants see as most consistent with the dual mandate given to it by Congress.” Note that Bernanke only commits to targeting longer run inflation forecasts—multi-year forward inflation rates, not the 12-month forward rate that I would like to see targeted.
At first the idea of using forecasts of inflation or NGDP growth as monetary policy indicators seems very odd; we are used to concrete policy tools like the Fed funds rate or the monetary base, not vague concepts such as expectations. But on closer analysis there is nothing that could be more sensible. After all, why should the Fed set policy at a level where they do not expect to hit their target? Why should they adopt a policy stance that is expected to fail? If they want 2 percent inflation, and their forecast unit expects 0 percent inflation, doesn’t that indicate that policy is too contractionary? And yet this is essentially what happened last fall. You might wonder how I can be so certain about this, after all, the Fed does not even set an explicit inflation target, so how can I possibly know that they expected to miss their target?
To my great surprise I found very few economists who shared my view that monetary policy was disastrously off course, far too contractionary for the needs of the economy.
Although the Fed has never set an explicit policy target, they have frequently hinted that they prefer roughly 2 percent inflation. One reason that the Fed has not set an explicit target is that important congressmen such as Barney Frank oppose the idea, and like to remind the Fed of its “dual mandate” to focus on both inflation and unemployment. For this reason, I suggest that they adopt a NGDP target, which takes both inflation and real growth into account. And indeed the Fed does focus on both variables in the short run. Thus, if inflation is right on target but output is falling, the Fed will usually ease policy; and if inflation is 2 percent and the economy is booming, the Fed will often raise the Fed funds target.
During early October 2008, virtually every economic forecast and every market indicator showed both inflation and real output falling far below the Fed’s implicit target. I spoke to a number of prominent economists at this time and there seemed to be little question that the Fed was very worried about the future trajectory of prices and output. This was the first time in 25 years that Fed policy clearly seemed far off course, not just in terms of my own policy preferences, but in terms of any plausible targets the Fed might have. So what went wrong? To my great surprise, I found very few economists who shared my view that monetary policy was disastrously off course and far too contractionary for the needs of the economy. Instead, all the discussion focused on the Bernanke and then–Treasury Secretary Hank Paulson’s bank bailout proposal, as if that could fix a problem of falling NGDP. To me this got things exactly backward; a monetary policy expected to produce sharply falling NGDP would seriously worsen bank balance sheets, making the crisis much worse. It was as if they were trying to bail water out of a boat without first plugging the hole in the hull.
Over the last year, the highly pessimistic forecasts of economists and financial markets during early October 2008 have come to pass. Both prices and output have fallen sharply, dramatically worsening the banking crisis. In his recent “60 Minutes” interview, Bernanke admitted that the early bailout cost estimates had been far too low, as the weakening economy made the banking situation much worse than expected. No one asked him why the Fed set policy last October at a level expected to produce falling prices and output, or why the Fed adopted a policy stance expected to fail. Indeed most of the criticism that did occur last fall focused on the Fed’s ballooning balance sheet, reflecting worries that it might lead to high inflation in the future. What explains this lack of criticism for the Fed’s failure to enact effective stimulus? The most common explanation I get from other economists (for their lack of attention to monetary policy) is that we were in a liquidity trap, and monetary policy had done all it could. I find this response bewildering, for many different reasons.
What explains this lack of criticism for the Fed’s failure to enact effective stimulus? The most common explanation I get from other economists is that we were in a liquidity trap, and monetary policy had done all it could. I find this response bewildering, for many different reasons.
During the stock market crash of early October 2008, the Fed funds rate was set at 2 percent, which is not the 0 percent which occurs in a liquidity trap. It is very possible that an additional 2-percent cut would not have been sufficient to prevent the recession from dramatically worsening, but that hardly explains why it was not attempted at that time. Then, on October 6, the Fed made one of the most significant errors in their 95-year history by beginning a policy of paying interest on bank reserves. This is a policy that the Fed had been planning for some time, and which other central banks have adopted, but the timing could not have been worse. It essentially created a liquidity trap-type environment, at an interest rate still well above zero. Previous liquidity traps, such as in the United States during the 1930s and Japan much more recently, occurred when nominal interest rates fell to zero on short term government debt, and then bank reserves and government debt became almost perfect substitutes. In that environment an increase in the monetary base may simply be hoarded by banks, and thus fail to boost nominal spending.
The policy of paying interest on reserves prevented interest rates from falling to zero, despite huge increases in the monetary base. Many economists were lulled into thinking policy was expansionary when they saw the base nearly double in late 2008. During normal times such a policy would have led to hyperinflation. But monetary theory is based on the assumption that money earns no interest, and thus normally is not a good substitute for interest-bearing securities. After October, however, reserves were like super T-bills, with a higher yield, more liquidity, and zero default risk. These reserves were more like securities than “money” in the normal sense of the term. The entire point of an expansionary monetary policy is to put more money into circulation than the public wants to hold. The attempt to get rid of excess cash balances boosts spending on goods, services, and assets. Now the Fed was deliberately preventing that sort of expansionary effect, by paying banks to hold on to the extra reserves. The justified the policy by arguing that it prevented market interest rates from falling below their target, what economists Robert Hall and Susan Woodward correctly called a “confession” of contractionary effect.
It is very possible that even zero interest rates would not have prevented a severe downturn in late 2008, but the Fed could have gone much further. Once they decided to pay interest on reserves, there was no reason why they could not make that rate negative, at least on excess reserves. A high enough interest penalty would have forced banks to sharply reduce their demand for excess reserve balances, which would have dramatically boosted aggregate demand. In the very unlikely event that all these excess reserves were then hoarded by the public, the Fed could have begun quantitative easing in October 2008. I very much doubt this last step would have been necessary, as demand for base money by the public (in the absence of 1930s-style bank runs), does not change very rapidly. The roughly $700 billion in excess reserves during late 2008 was more than enough to hit any reasonable Fed NGDP target. If an excess reserve interest penalty had forced this money out into circulation, it would have nearly doubled the amount of cash held by the public.
During fall 2008 I was stunned by the lack of criticism of Fed monetary policy by most macroeconomists. I could find almost no one who blamed the Fed for the sharp contraction, despite the fact that much of modern macro theory assumes that central banks, not commercial banks, determine the path of NGDP. But I also found a very interesting pattern; although almost no one agreed with my criticism of the Fed, they disagreed for three very different reasons. One group of economists argued that because we were in a liquidity trap, monetary policy had lost traction, and we now needed to focus on fiscal stimulus. Paul Krugman is probably the most influential proponent of this viewpoint. But I also doubted that this could fully explain the profession’s passivity—after all, the most popular money and banking textbook (by Frederic Mishkin) says that “Monetary policy can be highly effective in reviving a weak economy even if short-term interest rates are already near zero.” I have always believed this to be true; but I began to wonder whether I was the only one.
It turned out that lots of economists are dubious of the liquidity-trap argument against monetary policy effectiveness. And there is good reason for that skepticism. During 1932, Fed attempts to revive the economy were hamstrung by gold outflows, but when Franklin Delano Roosevelt took us off the gold standard in 1933 he was immediately able to sharply raise prices and output. And the Japanese case doesn’t fit either: the Bank of Japan seems to have a target of roughly 1 percent deflation, as each time they approach price stability they have tightened monetary policy. So we do not have any examples of monetary policy being unable to boost nominal aggregates under a fiat money regime. Instead, many conservative economists seem to have refrained from criticizing the Fed for entirely different reasons.
During fall 2008 I was stunned by the lack of criticism of Fed monetary policy by most macroeconomists. I could find almost no one who blamed the Fed for the sharp contraction.
Some conservatives argued that there was not much sign of deflation, other than a brief drop in the Consumer Price Index when oil prices crashed, and thus there was no need for a more expansionary Fed policy. Notice that this view conflicts with not just those on the Left, but even many moderate and right-wing economists, who agreed with Krugman that faster growth in aggregate demand was desirable, but doubted the effectiveness of fiscal stimulus. This second group, which includes figures like Robert Lucas, argued that monetary policy was still effective, and indeed was the best way to boost aggregate demand. So why hasn’t this group been more critical of Fed policy? Lucas argued that the Fed had already adopted a highly expansionary policy, specifically citing the large increase in the monetary base.
I have similar views to many monetarists about the monetary transmission mechanism, and the power of monetary policy in a liquidity trap. So how did I end up with such a different view of the Fed’s role in the crisis? I think there are two reasons. Most economists overlooked the problem of interest payments on excess reserves, and most economists have never fully accepted the radical view that policy must be forward-looking, that the Fed must always target the forecast. Instead, many economists seem to still hold to the old “long and variable lags” view of policy, the idea that the Fed should do some adjustments, and then “wait and see” for the lags to play out. But there is no lag between monetary policy actions and expectations.
On September 16, 2008, the Fed made one of its most costly errors ever. Immediately after the failure of Lehman Brothers, the FOMC decided to leave the target rate unchanged at 2.0 percent. The FOMC statement indicated that they saw the risks of inflation and recession as being roughly balanced. And yet on that day the economy had already been in recession for nine months, and the spread between the yield on five-year Treasury bonds and five-year indexed bonds (a commonly used proxy for expected inflation) stood at only 1.23 percent. They chose to ignore inflation expectations, and instead focus on the relatively high inflation rate during the previous year. By last fall and winter, the indexed bond market showed near-zero inflation expectations over the next five years. Admittedly, the indexed bonds spreads can be misleading, but the consensus forecast of private economists was also well below the Fed’s target; even the Fed’s internal forecast was well below 2 percent. Today the indexed bond market is signaling an expected inflation rate of roughly 0.5 percent over two years and 1.4 percent over the next five years. While we do not have a direct market indicator of NGDP or real GDP growth expectations, both economists’ forecasts and the implicit forecasts embedded in stock and commodity prices pointed to very slow growth in the period after August 2008. In plain English, this means that both economists and investors expected both monetary and fiscal stimulus to fail. Monetary policy should have been much more expansionary last October, and it should be more expansionary even today.
There have been a few other voices who have spoken out forcefully against the stance of Fed policy. James Hamilton is very knowledgeable about the Fed’s balance sheet, and was one of the first to notice the perverse effects of interest payments on bank reserves. Robert Hetzel has an excellent paper showing that policy was too contractionary during August and September 2008. Earl Thompson has shared my interest in forward-looking monetary policies, and has strongly criticized the Fed for ignoring deflation signals.
In 1989 I published a paper arguing the Fed should create and stabilize a NGDP futures market. For instance, if the goal is 5 percent NGDP growth the Fed would agree to buy or sell unlimited quantities of NGDP futures contracts at that price, and let each transaction trigger a corresponding open market operation. If they had done so in the middle of 2008, then 12-month forward NGDP growth expectations would have stayed at plus 5 percent during the fall and winter. Instead they plummeted to levels far below zero, dramatically worsening the financial crisis and creating the worst recession since the 1930s. This policy blunder also resulted in a massive fiscal stimulus and bank bailouts, which will balloon the deficit for years to come. None of this had to happen, even with the subprime crisis. Those loan losses were sunk costs, and would not have prevented an aggressively expansionary policy from boosting nominal GDP. This crisis should be a wake-up call to the Fed and to macroeconomists in general. We need a much more aggressively forward-looking monetary policy.
Scott Sumner is professor of economics at Bentley University and editor of The Money Illusion, a blog that focuses on monetary policy.
Image by Darren Wamboldt/Bergman Group.