June 16, 2009

Defining Deviancy

Governor Kevin Warsh

At the Institute of International Bankers Annual Meeting, New York, New York

In a seminal essay delivered about 16 years ago, Senator Daniel Patrick Moynihan offered a striking view of the degradation of standards in society.1 He observed that deviancy--measured as increases in crime, broken homes, and mental illness--reached levels unimagined by earlier generations. As a means of coping with the onslaught, society often sought to define the problem away. The definition of customary behavior was expanded. Actions once considered deviant from acceptable standards became, almost immaculately, within bounds.2

Society moves on, as it were. Well-meaning efforts are periodically made to treat its failings. But if these efforts prove less than successful, citizens and policymakers alike tend to grow increasingly accustomed to the unfortunate statistics. Every bit the reformer throughout his decades of public service, Moynihan seemed reluctantly resigned to society's construct: "In this sense, the agencies of control often seem to define their job as that of keeping deviance within bounds rather than that of obliterating it altogether."3

Given the financial crisis, deep contraction in the real economy, and extraordinary fiscal and monetary responses, I cannot help but wonder what constitutes deviance in economic terms in 2009 and beyond. What level of real economic output and unemployment is expected and, more important, accepted? And what level of volatility constitutes the "new normal"?4 As I will discuss, we must be wary of macroeconomic policies that--in the name of stability-- may have the effect of lowering trend growth and employment rates.

In Moynihan's framework, will we in the official sector be accepting of periods of significant financial and economic distress, however infrequent? That is, will deviancy be defined down with the understanding that a rare crisis is the price for dynamic, robust economic growth? Or will the official sector say, "Never again--not on our watch," and become less tolerant of deviations in economic and financial conditions? Under the mantle of reforming capitalism, will policymakers instead define deviancy up, and seek to guarantee stability in our economic affairs?

I suspect that, for a time, policymakers will be more attracted to this latter path. Stability is a fine goal, but it is not a final one. Long after panic conditions have ended, stability threatens to displace economic growth as the primary macroeconomic policy objective. But we must recognize that the singular pursuit of stability, however well intentioned, may end up making our economy less productive, less adaptive, and less self-correcting--and in so doing, less able to deliver on its alluring promise. This fate, however, does not have to be ours. The U.S. economy is capable, in my judgment, of delivering more.

The Growth Experiment5
This most recent boom and bust is not, as they say, our country's first rodeo, but it may turn out to be the most consequential since World War II. And, here, I am not just talking about the near-term peak-to-trough changes in growth and employment levels, which are likely to prove significant.

Policymakers are revealing new policy preferences and prescriptions--fiscal policy, trade policy, regulatory policy, and monetary policy, chiefly among them. Long after the official recession ends, the choices being made may significantly alter the contour of the U.S. economy. The harder question that remains is whether these changes will prove beneficial.

From the mid-1980s through 2007, U.S. real gross domestic product (GDP) growth averaged more than 3 percent per year, and was less volatile than in previous decades. The average unemployment rate was less than 5-3/4 percent, a full percentage point less than in the previous 15 years. Most notable was the realized acceleration in labor productivity in the mid-1990s.

The bipartisan, pro-growth policies that predominated during this period contributed meaningfully to these gains. Tax and spending decisions generally sought to expand the economic pie. Trade policies were aimed at opening new markets to U.S. products and services, and removing barriers domestically. Regulatory policies permitted failure, and relied in equal parts on capital requirements, regulatory standards, and, no less important, market discipline. As a result, businesses were well positioned to adopt new efficiency-enhancing technologies and processes to excel in the pro-growth environment. These policies helped drive significant productivity gains, and remarkable U.S. and global prosperity.6

I do not mean to suggest that the period that preceded the crisis was a golden age. It wasn't. Even during this seemingly enviable, secular period of prosperity, the U.S. recessions of 1990-91 and 2001 remind us that periodic, cyclical weakness occurs. The labor markets deteriorated during these downturns, and at great costs to many families and communities. But these periodic deviations were accepted, in some sense, by the body politic. There were lessons to be learned, reforms to be implemented, and public and private practices to be improved upon. But the broad outlines of the growth experiment forged on, and were judged by most to be largely a success.

Hence, the relative stability that earned the Great Moderation its enduring moniker is not, in my view, the greatest economic accomplishment of the past few decades. The acclaim should be assigned to the resulting economic boom that massively increased economic output and living standards. Maximum sustainable growth, not stability per se, was the predominant policy goal. Strong growth, of course, was made possible, in part, by the low volatility that marked the period.7 Firms choose to invest more, for example, in less volatile economic conditions. But, pro-growth policies were the coin of the realm.

But, as I remarked to another group of international financiers more than two years ago: "The Great Moderation…is neither a law of physics nor a guarantee of future outcomes. It is only a description--an ex post explanation of relative prosperity. If policymakers and market participants presume it to be an entitlement, it will almost surely lose favor."8 Well, warnings notwithstanding, it was treated like an entitlement by far too many in the private and public sector, and as a result, lost considerable favor.

The Panic of 2008
Has the experience of the last 20 months caused the findings of the vaunted growth experiment to be fundamentally revised? During this recent period of turmoil, the imponderable, or what was previously thought to be virtually impossible, happened with great speed and force and frequency.9 Asset prices plummeted, and market volatility reached its highest level in decades. Financial market functioning was deeply impaired across most asset classes and geographies. As a result, the U.S. economy endured a sudden stop. In the final quarter of last year and the first quarter of this year, private employment registered the largest two-quarter percent decline since the mid-1970s, and real GDP saw the most dramatic decline in a half-century.

These data are clearly indicative of significant deviance, and justifiably raise questions about the success of the growth experiment. Policymakers are rightly disposed to react, respond, and revisit the presumed record of accomplishment. In revising the historical record, however, we should not too hastily discount the preceding period of prosperity. We must avoid a classic case of what behavioral economists term "availability bias," when decisions made are influenced disproportionately by more recent events.

Ultimately, I will leave it to economic historians to assess whether the Panic of 2008 was more anomalous than the period of prosperity that preceded it. I believe that the categorization of recent events as deviant, ultimately, will depend on what happens next. That is, if policy changes cause future economic performance to suffer, then the boom of the last generation may, regrettably, turn out to be more exceptional than the bust.

The Stability Experiment
That new policies are being implemented during a period of economic turmoil is no coincidence. These new preferences reveal much about the type of economy to which policymakers aspire. And to be clear, the stability experiment appears well intended. It aspires to manage the economy with greater care and more expansive and effective regulation, as well as a larger and more persistent role for government action, and an increase in home bias in global commerce.

Advocates of the stability experiment--each guided by their own compass--seem all too inclined to announce that the growth experiment has ended, and to conclude that the results are deeply disappointing. They seem to prefer to define deviancy up, wishing to assure an uncertain citizenry. If government policies are corrected, and private practices are made more prescriptive, they argue, the ship of state can ensure that the real economy avoids rough seas altogether. The political economy confirms the policy response: Policymakers may pay greater attention to insuring against rare bad events--to which they could be held to account--than to allowing scores of great things to flourish.

As I mentioned at the outset, stability, in the sense of lower macroeconomic volatility, is a fine objective, but it is not a final one. The conduct of monetary policy, for example, aims to achieve price stability throughout the economic cycle. But central bankers do so because we believe it is the precursor to strong and sustainable growth. If other policies--notably, those for fiscal, regulatory, and trade--treat stability as the ultimate objective, then we might find ourselves with lower growth and diminished economic potential.

Reform efforts are sorely needed, and practices demand marked improvement. But policymakers are wise to resist the hubris that most recently afflicted actors in the financial sector.

I describe this mix of new pro-stability policies as an experiment for good reason. It might deliver on its promise of lower volatility, lower unemployment, and higher growth over the course of a generation. But, it might not. The costs of the stability experiment might turn out to be large. Necessarily and hastily crafted when the financial crisis began, the stability experiment is likely to survive far longer than the panic that preceded it. It is surging in popularity and is likely to grow in application, particularly if, as I suspect, the economic picture disappoints.

Incipient Recovery
Since mid-March of this year, financial conditions have continued to improve. Panic conditions are showing signs of retreat. Asset prices are rebounding, searching for a new equilibrium from their panic-induced depths--both in the United States and across the vast majority of our trading partners. And this improvement in markets coincided with the arrival of the proverbial green shoots of spring. This gloss of recovery is appealing, of course, and not only to central bankers.

I, like you, am rooting for the positive trend to continue. But, in my estimation, the rather indiscriminate bounce off the bottom--across virtually all assets and geographies--may be more indicative of a one-time reset, which may or may not be complete. I would be more comfortable going forward if we observe more dispersion in the valuations of particular assets and greater differentiation across asset classes and geographies.

The panic's hasty retreat should not be confused with robust recovery. For economic performance will turn ultimately on the force of private final demand; and for now, it remains weak. Real consumer spending rose only modestly in the first quarter of this year, after dropping sharply in the second half of last year. Businesses reduced real spending at an annual rate of more than 35 percent and continued to cut their workforces. Real exports fell at an annual rate of close to 30 percent.

The trauma experienced by businesses and consumers coming out of the panic should not be underestimated. Notwithstanding recent encouraging signs that the contraction is abating, I would expect business capital expenditures and consumer spending to continue to disappoint for the next several quarters. Even if, as I expect, the United States emerges from this recession sooner than our advanced foreign trading partners, I am still very cautious about predicting a sustained run-up in net exports so soon after the virtual collapse of global trade.

Exceptional fiscal expenditures, by their own terms, are intended to replace shortfalls in aggregate demand. And recent extraordinary monetary policy actions are intended to lower risk-free rates and grow balance sheet capacity to help offset the pullback by private financial intermediaries. But financial markets may extract penalty pricing if fiscal authorities are unable to demonstrate a credible return to sustainable budgets. And they are unlikely to look kindly on monetary authorities unless they decidedly and unambiguously chart their own independent paths. The Federal Reserve should not--and will not--compromise another kind of stability--price stability--to help achieve other government policy objectives.

On balance, I would not be surprised if these countervailing forces--unprecedented public support and underwhelming private demand--fight to a draw by the fourth quarter. But the scale of stimulus and the recent blow to the real economy are both lacking precedent, so predicting the victor is tough business. Of course, the extraordinary monetary and fiscal support may prove more efficacious in the near term than I expect, leading to a continued easing in credit conditions, a slowdown in the rate of deleveraging, improved inventory levels, and better quarterly economic statistics. Even so, the benefits of stimulus are likely to wane. More important, unemployment rates, in my judgment, are likely to remain higher and linger longer than in recent recessions. The "jobless recovery" may prove to be a familiar and vexing refrain. As a wise Stanford mentor of mine coined many years ago, "The economist's lag is a politician's nightmare."10

Medium-Term Prospects
The rebalancing of U.S. GDP and global demand is likely to require some patience. During the transition, there may well be political impetus for still more-aggressive macroeconomic policies. In evaluating new measures, however, policymakers' predominant interest should be ensuring the credibility of their fiscal and monetary frameworks. For, if macroeconomic policies were to become unanchored, or misunderstood by markets, continued government aggressiveness could prove counterproductive.

The global economy runs the risk of being mired in a period of slower growth for several years to come. Some portion of the subpar economic performance may be owed to the normal capital and labor reallocations that take place during recoveries. And given the serious misallocations that marked the onset of this recession, there is good reason to believe that the period of reallocation will be deeper and last longer. A reduction in the size of the finance and housing industries, for example, is well under way. Efforts to forestall those changes, in my judgment, are unlikely to succeed as promisingly as advertised. But perhaps a larger risk is that changes in public policies may, in the pursuit of stability, hold down the growth of the U.S. economy over this period.

Simple textbook models tell us that a nation's output is the product of the number of hours worked and the output produced per hour. In thinking about the medium-term prospects for the economy, I find it useful to consider these two factors separately.

Output per hour, or productivity, is the secret sauce to U.S. economic growth and to rising living standards, but I fear that the recipe may have lost some key ingredients. Growth in labor productivity arises when a firm's workers use more and better physical capital, or when firms become more efficient at converting inputs into output. Innovation plays a key role, both because it directly boosts efficiency and because firms' decisions to invest in physical capital tend to depend on the underlying pace of innovation. In addition, in today's economy, the productivity of many firms relies heavily on intangible, or intellectual, capital; although hard to measure, intangible capital appears to also be tied to innovation.

To be concrete, from 1995 through 2007, U.S. labor productivity growth in the nonfarm business sector averaged about 2-1/2 percent per year, a marked improvement from the 1-3/4 percent pace that marked the prior quarter century. This period of rapid growth in labor productivity was driven by large capital investments, significant improvements in management processes, and remarkable advances in technology. Looking ahead, if policy is less encouraging of capital accumulation, or returns to innovations are constrained by policy, we may find a material reduction in the growth rate of productivity and living standards.

The level of support that the financial system is capable of providing also remains highly uncertain. Private financial institutions are now understandably slow to create new products that connect savers and investors. Although it is undesirable to revert to the excessive risk-taking that preceded the crisis, current financial practices seem suboptimal in promoting economic growth. Furthermore, repeated interventions by the public sector run the risk of causing systemically significant institutions to operate more like public utilities than efficient allocators of capital and proper arbiters of liquidity.

Productivity may also suffer at the hand of policies that discourage trade. Trade enhances productivity by promoting efficient specialization, permitting economies of scale, and increasing the potential returns to innovation.11 However, the bipartisan consensus favoring free trade appears broken, with each political party internally divided on the question. Given the contention, however imprecise, that Anglo-American-style capitalism caused the turmoil, there may be a shortage of credible and persuasive voices to fight the growing global tide of economic isolation.

As for hours worked, the other key determinant of output, the risks seem unmistakable. We have grown accustomed to falling natural rates of unemployment during the last two decades. Even when unemployment spiked during recent recessions, the prevailing rate found its way to still lower levels. According to most economists, the non-accelerating inflation rate of unemployment (NAIRU) was in the neighborhood of 5 percent prior to the financial crisis.

Looking ahead, I could well imagine that the natural rate of unemployment trends higher. Historically, small businesses have tended to be the largest drivers of new job creation. But how they will respond if macroeconomic policies favor stability over growth is difficult to predict. The answer may ultimately depend on their access to growth capital and liquidity given the changing mix of public policies and private practices.

More generally, I believe that the dynamism of the U.S. economy contributed critically to pre-crisis levels of employment. The remarkably low level of the NAIRU was made possible in part by the extraordinary churn in jobs seen in the U.S. economy. For example, over the 12 months preceding the start of the recession in December 2007, although 62 million people lost their jobs, 63 million jobs were created. Policies aimed at limiting the range of economic outcomes do not, in my view, keep unemployment rates low. These policies make it more costly for firms to adjust the size of their workforces, thereby making employers more reluctant to hire.

Gauging trend growth in output and employment is essential to the proper conduct of monetary policy. Broad changes in macroeconomic policies may make it more difficult for central bankers to make such calculations. To the extent changed policies reduce potential growth, or raise the natural rate of unemployment beyond recent estimates, it is more difficult to make good and timely policy. We must keep a keen eye on these risks as conditions evolve.

Ironically, the longer that output and employment disappoint, the more attractive the gloss of stability may become. As a result, we should evaluate the benefits and costs of the stability experiment with a keen eye to guard against downside risks for the real economy.

Longer-Term Prospects
I have not lost confidence in the inherent innovation, creativity, and dynamism in the U.S. economy. Nor have I lost confidence in the inherent good sense of our citizens. If the stability experiment fails to deliver on its promise of higher employment and better economic performance, then policymakers ultimately will change their prescriptions yet again. And so, in the long term, after the final results of the U.S. experiments with growth and stability are finally tabulated, the broader U.S. experiment in democratic capitalism will endure--and our economy will emerge stronger than ever.


1.  Daniel Patrick Moynihan (1993), "Defining Deviancy Down:  How We've Become Accustomed to Alarming Levels of Crime and Destructive Behavior," American Scholar, Winter. Return to text

2.  The views expressed herein are my own and do not necessarily reflect the views of other members of the Board of Governors or of the Federal Open Market Committee.  I am grateful for the valuable assistance of Karen Dynan and Nellie Liang of the Board staff who contributed to these remarks. Return to text

3.  See Moynihan, "Defining Deviancy Down," note 1. Return to text

4.  See Mohamed El-Erian (2009), "A New Normal," Pimco Secular Outlook, May.  Return to text

5.  The term "Growth Experiment" was coined two decades ago to describe certain pro-growth tax policies.  See, for example, Lawrence B. Lindsey (1990), The Growth Experiment:  How the New Tax Policy Is Transforming the U.S. Economy (New York:  Basic Books).  I, however, am referencing a broader range a macroeconomic policies. Return to text

6.  Since 1980, the performance of the world economy has proved strong.  Real world GDP grew by approximately 145 percent from 1980 to 2007, a compound average of about 3.4 percent per year (Becker and Murphy (2009)).  They also contend that even if the current recession were to be deeper and more prolonged than contemplated by the most pessimistic forecasters, global annual growth rates would still average better than 2.7 percent from 1980 to 2010, with per capital incomes rising a total of about 40 percent over the period.  See Gary Becker and Kevin Murphy (2009), "Do Not Let the 'Cure' Destroy Capitalism," Financial Times Magazine, March 19. Return to text

7.  Whether by good fortune or good policy, between the start of the Great Moderation in the mid-1980s and the end of 2007, the standard deviation of annualized quarterly real U.S. GDP growth was 2.1 percent, less than half as large as its value over the two previous decades.   Return to text

8.  Kevin Warsh (2007), "Market Liquidity:  Definitions and Implications," speech delivered at the Institute of International Bankers Annual Washington Conference, Washington, March 5. Return to text

9.  Kevin Warsh (2009), "The Panic of 2008," speech delivered at the Council of Institutional Investors 2009 Spring Meeting, Washington, April 6. Return to text

10.  See George Shultz (1993), Turmoil and Triumph:  My Years as Secretary of State (New York:  Charles Scribner's Sons). Return to text

11.  For a discussion of the benefits of trade in the modern global economy, see W. Michael Cox and Richard Alm (2007), "The Best of All Worlds: Globalizing the Knowledge Economy," in Federal Reserve Bank of Dallas: 2006 Annual Report (Dallas:  FRB Dallas). Return to text

Last Update: June 16, 2009