Why the Experts Missed the Recession

The interest rate policies for the US are set in secret by the Federal Open Market Committee (FOMC) at regularly scheduled meetings eight times a year. Occasionally, the chairman of the FOMC, who is also the chairman of the central bank, the Federal Reserve, holds emergency conference calls as well. The decisions of the committee are briefly reported the next day and only a cursory summary of the minutes of the meetings are published a few weeks later. But a full transcript of the forecast and discussions is not published for five years. Earlier this year, the Fed published the staff’s economic forecasts and the full transcripts of the meetings during the crisis of 2008. The result is a fascinating and disturbing account of how respected experts got things wrong.

By lowering the target rate of interest, known as the federal funds rate, the members of the FOMC can stimulate economic growth, and by raising it, they can dampen growth and inflation. Such Federal Reserve activity, which is accomplished, for the most part, by buying and selling federal securities, has been the dominant policy instrument for controlling the pace of economic growth and inflation in the US since the 1980s.

The year 2008 was momentous economically, and the published proceedings show one of the nation’s most important policymaking bodies operating mostly in the dark about the impending crisis. America and Europe suffered the worst recession since the Great Depression of the 1930s. Unemployment in the US soared to 10 percent, and in some European nations to 25 percent, and millions here and abroad lost their homes. The newly published record shows an extremely limited state of economic knowledge and forecasting ability on the part of committee members. It also shows how undependable attempts to adjust the economy in the short term can be.

Like most economists, the Fed policymaking committee not only failed to anticipate the severity of the Great Recession, it was unaware even that the moderate recession that began in 2007 was underway until it was nearly one year old. “The Fed’s inability to forecast…is disconcerting,” wrote one commentator on the record. But many studies have shown that neither government nor private economists can forecast the economy with any consistency. While they often anticipate the direction of the economy, such forecasts have only limited success because the economy is growing 80 percent of the time. Predictions about when it will stop growing are almost universally inaccurate. Among the studies of the general forecasting record of economists, one of the broadest was recently undertaken by Hites Ahir and Prakash Loungani, both of the International Monetary Fund. They found that, based on a survey of forecasts in many different countries, not one official or private forecast in 2008 anticipated a recession in 2009. Yet there were recessions in forty-nine countries that year.

A recession is defined as a prolonged decline in national income or production—a drop in Gross Domestic Product. In the US the customary definition is that a recession occurs when GDP falls for two calendar quarters. In such a period, unemployment rises, sometimes sharply, wages often stagnate, bankruptcies are frequent, and business investment weakens. Since World War II, such recessions have been moderately frequent but usually of short duration.

It is one thing to fail to anticipate a moderate recession, however, and quite another to miss one as devastating as what became known as the Great Recession of 2008. The US recession began at the end of 2007 and lasted through the middle of 2009. GDP fell more than 5 percent overall, and the number of people employed as a percentage of the working-age population plunged. New records were set for those unemployed for more than six months. Millions of mortgage holders now owed more to the banks than their homes were worth, a result of the boom in mortgages issued to those with poor credit. And those bad mortgages had been in turn incorporated into financial instruments such as “collateralized debt obligations” that were irresponsibly sold by American banks and others throughout the world, causing heavy losses to large and small investors. Making matters worse, the recovery in employment has been historically slow since then.

In “Can the Fed Predict the State of the Economy?,” Tara Sinclair, who is codirector of the forecasting group at George Washington University, compared a consensus of private forecasts, regularly compiled by the Federal Reserve Bank of Philadelphia, to the actual results during the Great Recession. In mid-2008, the consensus forecast anticipated that there would be eight million more jobs in 2009—some 6 percent of the work force—than there actually were. It was similarly wide of the mark for GDP and inflation.

The forecast of the Fed’s Open Market Committee was no different. For each meeting the Fed’s economics staff, arguably the best-trained in forecasting today, prepares both a summary of current conditions and forecasts of up to two years. These are known as the Greenbooks, based on the traditional color of the bindings. Researchers have compared the Greenbook forecasts to the private forecast consensus and find that they are not superior. Sinclair and her coauthors find, in fact, that the Fed Greenbook at best tells us the state of the economy in the current quarter but cannot reliably predict the state of the economy even one quarter ahead.

The recently published set of 2008 FOMC transcripts and Greenbooks amounts to a classic account, unrivaled by any book that I’ve read, of how little most economists knew about what was happening during the Great Recession. Even as late as the spring of 2007, Fed Chairman Ben Bernanke argued that a major disruption due to a collapse in subprime mortgages could be contained.

By late 2007, Bernanke was changing his mind, as were other members of the FOMC. Mortgage holders were defaulting on their payments in rapidly growing numbers and mortgage brokers were going out of business. Beginning in August 2007 Bernanke urged cuts in the Fed’s target rate—the “fed funds rate,” which is the rate banks charge each other for overnight borrowing to meet federal requirements. By early 2008, however, the economy was weakening measurably. The unemployment rate had risen in 2007 and while GDP kept growing, there were visible strains on banks.

In January 2008, the FOMC cut the target rate by half a percentage point, but the economy now started to soften further, as consumer spending weakened. Bernanke and the FOMC continued to cut rates through April. The fed funds rate at that point was 2.0 percent—a substantial easing under normal circumstances from 4.25 percent a few months earlier.

But these were far from ordinary times. The Greenbooks make disturbingly clear that Bernanke and the FOMC did not understand the implications of the economy’s fragility until the autumn of 2008, and even then they badly underestimated the devastation that was underway. To the contrary, there was constant concern that the FOMC’s policies of easing interest rates might reignite inflation. As Jeffrey Lacker, president of the Richmond Fed, said at the September 2007 meetings, “I’m concerned about the likelihood of inflation’s remaining above my objective.” Several members of the FOMC consistently sounded the same note, as if a small amount of inflation would be a natural disaster, not a possible impetus to parts of the economy.

If anything becomes devastatingly clear in reading the Greenbooks and the FOMC response, it is that the Fed was overly quick to respond to positive turns in the monthly information. In March 2008, the Fed had sensibly helped to save Bear Stearns, the fifth-largest investment bank, which had gorged itself on bad mortgage securities, by selling it to JPMorgan Chase. The sale calmed the markets. In the second calendar quarter, GDP strengthened as consumer spending reversed its first-quarter weakness. Now some on the FOMC thought its interest rate cuts had saved the day. At an FOMC meeting toward the end of April, the Fed governor Fred Mishkin, a former Columbia professor, said he had been gloomy early in the year. Now his views had happily changed. “I think,” he said, “it is very possible that we will look back and say, particularly after the Bear Stearns episode, that we have turned the corner in terms of the financial disruption that we have just experienced.”

The other FOMC voting members were satisfied to stop easing the interest rate after April and left the fed funds rate unchanged at 2 percent through the September meeting. Even a hint of optimism invariably led to warnings about renewed inflation from some of the FOMC participants. Dallas Fed President Richard Fisher was persistent in such fears of inflation. In the June 2008 meeting, he ominously told his colleagues that “the real bad news is that our patient appears to be acquiring a staph infection in this hospital that we have created, and that staph infection is inflation.” He couldn’t have been more wrong. Inflation has still not risen from levels below 2 percent a year. Fisher wanted to raise the fed funds target throughout the summer of 2008.

In light of improving conditions—or so it seemed to forecasters—the Fed and the Treasury decided not to save Lehman Brothers later that summer. Lehman was much bigger than Bear Stearns and therefore a greater risk. The FOMC believed, despite falling stock prices, that the financial markets and the economy could now absorb a bankruptcy, the largest ever in America to date. Some may have worried about creating a “moral hazard,” in which investors feel safe to take undue risks knowing the federal government will bail them out, a common concern especially of more conservative policymakers.

Contrary to the beliefs of the FOMC, the Lehman bankruptcy immediately led to financial panic. Bankers felt that they had to sell securities to meet obligations to one another, driving the value of securities down further in a vicious spiral. European markets also were failing. The leading money market fund, the Reserve Primary Fund, found itself stuck with Lehman Brothers securities that had gone bad; the fund could not pay investors their money back in full. A run on the money market funds, thought as safe as a federally insured savings account by investors, was now possible. And the stock market was plunging.

The Greenbook’s forecast prepared for the September meeting the day after the market panic began, however, still did not anticipate a serious recession. It predicted that GDP would grow at an annual rate of 1.1 percent in the fourth quarter, only a few weeks away. As for the longer run, the September Greenbook projected that “the pace of real GDP growth will step up to about 2 percent in 2009 and to 2¾ percent in 2010.”

The prospect of a steep recession was not on the committee’s mind. Some, like San Francisco Fed President Janet Yellen, now the Fed chairman, were concerned that the economy could weaken more rapidly; so was Eric Rosengarden of the Boston Fed, and Bernanke himself. But at the beginning of the worst collapse in economic performance in seventy years, the Fed decided to leave the fed funds rate unchanged.

Within a few weeks, this outlook changed sharply. The financial markets had been falling fast. Early in October 2008, in an emergency phone conversation, the FOMC cut the fed funds rate to 1.5 percent. In its regular meeting in late October, the staff reported that “recent economic and financial news has been dismal.” At last, the October Greenbook anticipated a recession, with a moderate fall in GDP in the third quarter and a decline at an annual rate of 1.25 percent in the fourth. It is hard to fathom how wrong the Fed forecasters still were.

It was the financial crisis, not economic forecasts, that forced the Fed’s hand. Under Bernanke’s leadership, the Fed provided a variety of credit guarantees to supplement the $700 billion provided in the Troubled Asset Relief Program (TARP) passed by Congress to supply funds to banks and other companies. Even at this point, the Fed economists were forecasting a fairly early economic bounce-back in 2009. The unemployment rate might run as high as 8 percent or so, they thought, but then it would recover. In October, the Fed cut the fed funds target rate only half a percentage point to 1.0 percent. Over the next six weeks or so conditions deteriorated still more rapidly. In early December, the Greenbook now forecast a drop in fourth-quarter GDP of an annual rate of 4.7 percent, a stunning shift from its moderate optimism only two months earlier. But even this sharp revision was far off the mark. It turned out that the economy fell at an annual pace of slightly more than 8 percent that quarter.

The Fed had undertaken another special emergency measure by buying hundreds of billions of dollars of mortgage securities the month before, in November. This helped keep interest rates low, supplementing further cuts in the fed funds rate. In December, the Fed basically cut the fed funds rate to zero. It now could go no lower, but the Fed continued to make more funds available by buying public and private securities, a policy known as quantitative easing.

Poor forecasting also led the White House to make a major political blunder. Soon after taking office in early 2009, Obama’s advisers wisely urged him to back a fiscal stimulus—a combination of government spending and tax cuts—of roughly $800 billion. The aggressive stimulus shortened the recession and got the economy growing again by the summer. It was a bold initiative, if not bold enough. In my own view, Obama could have asked for a second stimulus but he did not raise the issue again until his second term, so strong was the political antagonism toward federal budget deficits, particularly in Congress.

Unfortunately, the Obama forecasters were also dangerously overoptimistic about the unemployment rate. They promised the public that unemployment would not rise above 8 percent, much as the FOMC staff had anticipated. The White House economists had correctly predicted how much the stimulus could encourage growth but had not realized how big a hole the economy had already fallen into. In fact, the unemployment rate topped out at 10 percent in the fall of 2009, and fell only slowly over the next four years. Obama’s unfulfilled promise of a limit of 8 percent was exploited by his political antagonists. Bad forecasting cost Obama dearly.

In retrospect, there are explanations for why the forecasting failed so badly. The Fed staff and the FOMC members were well aware after the Lehman bankruptcy that financial markets were panicking. The financial markets were at this point, however, far more fragile than was recognized. The FOMC and staff were also aware of the turmoil underway, but they could not anticipate what would happen next. Money market funds, commercial paper, and the markets for derivatives, which were mostly unregulated, were all caught in a downward spiral, investors afraid their creditors would not pay them back. Derivatives were the highly leveraged securities based on stocks, bonds (including mortgages), and currencies that now dominated the markets. The Fed staff seemed to have insufficient knowledge of how risky in particular the mortgage securities were. As investors across the board sold securities, they made matters worse by driving down prices further, forcing investors with debt obligations to sell more.

There was a still greater forecasting problem, however, one to which the Fed staff readily admitted. They had little experience with how a financial crisis of this dimension could affect the real economy of production, jobs, wages, and profits. As the Greenbook stated in somewhat technical language:

The infrequency of episodes of financial turmoil, difficult econometric identification problems, and the absence of a generally accepted paradigm for real-financial linkages suggest that our standard models probably do not fully capture the effects of changes in financial conditions on real GDP.

This failure reflected a major fault line in economic theory. Although highly respected economists like Hyman Minsky and Charles Kindleberger had documented the impact of financial market excesses on economies back in the 1970s and 1980s, contemporary economists by and large believed that any such imbalances in financial markets were modest and short-lived. In their view the workings of the invisible hand of the market would set prices that were on the whole accurate and efficient. Low inflation was basically all that mattered.

With few exceptions, macroeconomic theory simply treated financial markets as a harmless sideshow. Financial regulation to ameliorate excesses was not considered part of economic management. “We thought of financial regulation as mostly outside the macroeconomic policy framework,” conceded former MIT economist Olivier Blanchard, now chief economist of the International Monetary Fund. As it turned out, little could be further from the truth.

Dan Sichel, a former senior member of the Fed staff, now a professor at Wellesley College, thinks the sharp slide of the real economy beginning in the late summer of 2008 makes it clear that the credit collapse that followed Lehman’s bankruptcy led to the steepness of the economic plunge. As he says, companies had no access to credit and cut back on activities and employment. The Fed staff did try to account for potential financial disarray in the months before the late summer of 2008; but even their most dire hypothetical forecasts did not come close to the mark.

With a better forecast, the Fed might have reduced rates faster. It might even have considered the then-unthinkable—raising its inflation target considerably from its customary annual rate of 2 percent. A higher inflation rate would have effectively lowered some interest rates below zero, encouraging savers and businesses to spend and invest rather than put their money in the bank and lose purchasing power. The interest rates would not have been high enough to offset inflation.

Ironically, Bernanke was well aware of how a collapse in credit could lead to serious consequences, lessons he learned as a scholar of the Great Depression of the 1930s. But he was oblivious to the extent of the risks until it was late in the day. One limitation on the available knowledge was that there were too few historical examples to learn from.

But knowing that a credit collapse caused the severity of the recession does not mean that simply reversing it—that is, making the banking system solvent again—would lead to recovery; and this was another mistake made both by Bernanke and by George W. Bush and Obama administration officials. Their goal was to make the banking system sound by supplying it funds from TARP and guaranteeing other lines of credit to forestall runs on money market funds and commercial paper.

Unfortunately, supplying funds did not mean the banks would in turn lend the money; nor did it mean that businesses and consumers would want to borrow and spend it. A more effective bank rescue could have required higher levels of lending under, in some cases, new bank management, but there were no such requirements. Nor was there new management; in fact, no bank CEOs were dismissed. Some economists suggested outright nationalization until the banks became healthy and more proactive. Still others suggested that banks be broken up to reduce their power and propensity for risky investments, and dilute the adverse financial consequences should any of them go bankrupt.

Not all economists agreed that the panic, however, was the main cause of the severe recession. Rather, they called for a more targeted set of policies to reduce sharply the indebtedness of households, a policy that still remains to be adequately carried out. A perceptive book, House of Debt, by Atif Mian and Amir Sufi, makes a strong case that the excessive level of borrowing by middle-class and even poor Americans was a fundamental cause of deep recession. Once unemployment started to rise, Americans had less buying power because they were strangled by mortgage and other debt.

Even a Fed policy of cutting interest rates earlier and faster, the authors contend, would probably not have been adequate to offset the loss of aggregate buying power. American homeowners needed sharp reductions in their monthly payments. One way to do this could have been to force banks to reduce the principal owed on mortgages to a level closer to the new lower market values of the homes. Another possibility was to liberalize bankruptcy laws, enabling homeowners to negotiate lower payments. Ironically, the Obama administration did have authority to take some of these actions. But the Obama economics team was loyal to a strategy of making banks strong, and not weakening them by reducing the value of the loans they had made.

For all their data, Mian and Sufi may have oversimplified the case. The nation would have had a serious recession based on its high levels of debt, but it is hard to argue that the severity of recession would have been nearly as great without the collapse of credit in the fall of 2008. The unemployment rate rose from 6.1 percent to 7.2 percent in just the four months following the Lehman bankruptcy, and then kept on going. Industrial production fell in a straight line from mid-2008.

One of the constructive contributions made by Mian and Sufi has been to restore awareness of how important consumer buying is to economic growth. They also argue that classic Keynesian stimulus was needed to bolster national buying power. Bernanke himself agreed. But after the beneficial $800 billion stimulus the Obama administration pushed through Congress in 2009, the nation got no more. Instead there was a hesitation by the government to spend and then an outright reduction of government spending in 2013 under the congressional sequestration, a policy the Obama administration agreed to.

Meanwhile, economists are trying to adjust their forecasting models to take account of the financial sector and its many potential distortions. A survey of these efforts by Serena Ng and Jonathan Wright—in their report “Facts and Challenges from the Great Recession for Forecasting and Macroeconomic Modeling”—is not encouraging. The unknowns seem overwhelming, and the predictability of the many variables is so far limited.

The Fed is again facing major decisions under Janet Yellen. The persistent inflation hawks are sounding their concerns that the nation should be raising interest rates now and cutting back on quantitative easing sooner rather than later. Others are urging the Fed to adopt a rule to set monetary policy based on a combination of inflation, employment, and GDP growth rates.

The most prominent is known as the Taylor Rule, devised by the former Bush administration economist John Taylor of Stanford. But such a rule also requires interpretation, such as setting inflation target rates and making assumptions about when the nation is at full employment. Taylor’s own interpretation of the rule requires a higher target for the fed funds rate today. Yellen’s interpretation gives more weight to high unemployment and she opposes higher fed funds rates as a result, though she may favor regulations that temper speculation in, for example, junk bonds, the high-yield bonds of risky corporations, or relatively new subprime auto loans. Judgment calls simply can’t be avoided.

Under Yellen, the Fed is likely to take a longer perspective and not react immediately to a month’s good news. This would be highly welcome. But the other members of the FOMC will publicly sound their traditional concerns, many of them ideological. Sadly, the science of economic forecasting is not nearly sound enough to subdue such ideological biases. The inflation hawks have been dead wrong for so long that one would think they would be more cautious.

If we expect the Fed to shoulder the entire responsibility for renewed growth in America, even under Janet Yellen, we will be disappointed. A stronger safety net to help the poor and significantly more public investment in infrastructure, early education, and, I’d argue, the relief of child poverty are requirements for prosperity. Yellen will face many antagonists. The Fed deserves some blame, but not all of it. Economic forecasting is too unreliable to silence errant, irresponsible voices.