By Richard Alford, a former economist at the New York Fed. Since then, he has worked in the financial industry as a trading floor economist and strategist on both the sell side and the buy side.
“..there is always a well-known solution to every human problem — neat, plausible, and wrong.” H L Mencken
One Tool One Target
US Policymakers and elected officials are again debating the most appropriate policy mandate for the Fed. The two principle contenders are the dual mandate (inflation and unemployment) and the inflation-only mandate. The proponents of these two mandates present them as “optimal” or at least superior to any alternatives. However, the inflation of the late 1970s and the economic and financial dislocations that followed the recent asset price bubbles demonstrate that policy driven by Phillips Curve considerations alone can contribute to less-than-satisfactory outcomes. Nonetheless, economists and policymakers appear to be unable to divorce themselves from the Phillips curve mentality, which posits that there is a tradeoff between inflation and unemployment.
During the oil shocks of the 1970s, the Fed attempted to insulate the real economy and the labor market from the repeated OPEC-induced negative supply shocks. This effort resulted in a serious acceleration in the rate of inflation and behaviors of inflation and unemployment that were inconsistent with the Phillips Curve hypothesis. The return path to price stability ran through a costly double-dip recession.
The experience of the 1970s ended the use of the Phillips Curve as a guide to policy. The “natural rate of unemployment hypothesis” replaced the Phillips Curve in economists’ tool kits. The natural rate hypothesis posits that while macroeconomic policy could have short-run effects (via transitory inflation-augmented Phillips Curves) in the long run the unemployment rate would return to its natural rate (no long-run trade-off between inflation and unemployment). Any short-run effects on output and employment are viewed as the result of the deviation of actual inflation from expected inflation.
Post-1996, especially post-2001, this recast Phillips Curve/Taylor Rule mentality failed to deliver sustainable growth. The underlying problem was in part an external positive supply shock. Globalization and the inability of the Dollar to adjust contributed downward pressure on US price indices. At the same time, externally produced goods replaced domestically manufactured goods here and abroad. This was reflected in a widening US trade deficit and downward pressure on US employment, prices and output.
The Fed, adhering to its chosen variant of the Taylor Rule, eased policy in response to both the-inflation and output shortfalls relative to target. Policy stimulated debt-financed consumption and real estate investment booms. Debt grew relative to income. Low costs of carry encouraged financial institutions to grow their balance sheets relative to their capital as they increased the size of their maturity mismatches. Despite the growth of debt, leverage, maturity mismatches and asset prices out of line with historical experience, financial stability considerations and the unsustainable external imbalance were dismissed as inflation and output/employment remained below target.
However, no steps were taken to address the external imbalance. Instead, policy masked the effects of the external deficit and the inability of the Dollar to adjust via the unsustainable asset price bubbles, which in turn led to unsustainable patterns of investment and consumption. When the unsustainable asset price rolled over, the US economy experienced the worst recession since the Great Depression, and a total collapse of the financial system was narrowly avoided. The external imbalances, however, are a larger problem than they were 10 years ago.
Numerous economists and policymakers outside of the US policy circles view the continued US attachment to the one-tool one policy target as anachronistic. At a recent IMF-sponsored conference entitled “Macro and Growth Policies in the Wake of the Crisis”, Olivier Blanchard, Director of Research at the IMF, described the pre crisis consensus that was held by economists and policymakers as the belief that monetary policy should use one tool (the policy rate or policy rule ) to pursue one target (price stability.) Post the crisis, Blanchard sees recognition of the fact that the policy problem is one of numerous targets (e.g. price stability, full employment, financial stability) and multiple tools (monetary policy, financial regulation, fiscal policy, capital controls). In short, Blanchard sees economic policy as more complex and messy than the pre-crisis consensus (see here and here for additional commentary on the conference)
Nonetheless, US policymakers and many economists continue to adhere to the one tool-one target approach. Despite the fact the Inflation-only target and the Taylor Rule constructs are ill-suited as guides to monetary policy not only in the face of supply and external shocks, but also due to the complexity of the interplay among the various policy tools and policy targets.
The continued attachment is not surprising. A new policy consensus is yet to emerge and specialization and the division of labor are at the heart of economics. However, Adam Smith acknowledged limits to specialization/division of labor even as he touted the potential benefit of specialization, e.g. the pin factory. He wrote: “…the extent of this division must always be limited…by the extent of the market. When the market is very small, no person can have any encouragement to dedicate himself entirely to one employment, for want of the power to exchange all that surplus part of the produce of his own labour…for such parts of the produce of other men’s labour as he has occasion for.” In short, it doesn’t pay for an economic agent to specialize unless some other agent or agents produce and are willing to exchange their surplus output.
Translated to the current policy context, it may be counterproductive for the Fed to single-mindedly pursue price stability given the absence of other policies that can effectively insure financial stability, external balance, etc., thereby insuring full employment and stable growth.
Adam Smith’s observation has a parallel in the literature of general equilibrium and theoretical welfare economics. It is known as the theory of the second best. This theory is concerned with the implications of having at least one optimality condition which cannot be met. The principle conclusion is that when at least one optimality condition cannot be satisfied, it is possible that the best achievable solution involves other variables/conditions taking on values other than those that would be optimal in a perfect world.
Placed in a macroeconomic setting, this suggests that the monetary policy that would be optimal in a perfect world may be inappropriate given the current moral hazard incentives, the inability of the exchange rate to adjust to maintain external balance, too-big-to-fail-financial institutions, the failure of financial regulation, etc. This is especially true given the ample evidence that expansionary monetary policy operates by increasing the risk profiles of financial institutions and the fragility of the financial system.
The idea that the optimal macroeconomic policy will be in part determined by the presence of market and organizational failures has also been noted in the literature. One of Jan Tinbergen’s (co-winner of the first Nobel Prize in economics) major contributions was to show that a government with several economic targets — e.g., the unemployment rate, the inflation rate, fiscal balance, financial stability and external balance — must have at least as many effective policy instruments to be assured of hitting all the targets.
There is a corollary: if the number of tools is less than the number of targets, policymakers will be faced with policy trade-offs. The corollary does not imply that a central bank or other policy maker cannot pursue one goal despite obvious deterioration of performance vis-à-vis other goals. The Fed did just that. However, It implies that society might have been better off with a larger miss relative to the inflation and output targets in 2001-2006, but with sustainable growth, a more robust financial system and a smaller external deficit. This begs the question:
How does the number of policy targets compare to the number of effective policy tools in the US today?
Policy targets (a partial list):
• Full employment
• Financial stability
• Fiscal balance
• External balance
• Stable growth
• Monetary policy (short-term interest rate or quantity of reserves), but perhaps not at the zero bound
Ineffective to Nonexistent:
• Fiscal policy — Are expenditure and tax policy (the tax code) effective policy tools? Are they designed and executed with policy goals in mind or are they the outcome of a problematic political sausage factory with only a passing resemblance to the policy goals cited above.
• Regulatory Policy The prior regime was ineffective and the new regime isn’t qualitatively different. The new regime is also the product of domestic and international sausage factory.
• Exchange Rate policy-Capital controls Nonexistent
• Trade policy-Nonexistent
The number of policy goals exceeds the number of effective policy tools. Given that significant misses of any of the policy targets listed above are inconsistent with the ultimate policy goals-(stable growth with full employment), the Tinbergen perspective implies that policy trade-offs should be expected. “Specialization” of policy tools is a mistake, especially if it contributes to larger policy misses elsewhere.
While the pre-crisis consensus is now viewed as “wrong” by many, a new consensus is yet to emerge. However, it appears that most US economists are in favor of marginal changes: incorporate a more realistic financial sector into their models, ignoring other factors such as:
1) the unsustainable external imbalance and the inability of the Dollar to adjust,
2) unsustainable patterns of consumption and savings relative to income (partially policy induced),
3) insufficient non-real estate real investment relative to income and the need to generate income for future retirees.
Blanchard offered opinions about the future of policy given the absence of a consensus. Loosely paraphrasing his presentation at the conference he said given the limits of our knowledge, policymakers must move slowly. They must not give up inflation targeting, but experiment moving step by step, altering targets and instruments one by one. It is unclear why Blanchard believes that policymakers have the luxury of moving slowly or why he believes that the global economy will be a well-behaved back drop while policymakers run their experiments.
His conclusion s about post-crisis policy would have been closer to the mark if he had said that policymakers should:
1) move at deliberate speed but cautiously taking into account possible undesirable “side effects” due to market and institutional failures,
2) acknowledge that policy tools do not map one for one into policy target, and
3) recognize that policy trade-offs are inherent given the number of effective tools and the number of policy targets
Policy and Irony
Mention of the global imbalances in Blanchard’s discussion of crisis was conspicuous in its absence—very ironic given he is associated with IMF and that IMF sponsored the conference. It is as if the discussants thought that a more robust regulatory system would necessarily resolve the global imbalances or the recent crisis was completely independent of the global imbalances.
The Tinbergen analysis presumes the existence of policy trade-offs when the number of policy targets exceeds the number of effective policy tools. The Phillips Curve acknowledges a single trade-off for macroeconomic policymakers. The Natural Rate hypothesis allows for none (at least in the long-run).
If nothing else, the absence of policy trade-offs in the current economic thinking is most ironic. By definition, economics is the study of trade–offs: the allocation of scare resources across competing uses, e.g., guns versus butter, labor versus capital, labor versus leisure and present versus future consumption.
However, US macroeconomists and the Fed have decided to side step the problem of the relative scarcity of macro policy tools. They have defined or assumed away all but one use for monetary policy. They continue to treat the optimal stance of monetary policy as if it were separable from other policies as well as from market and institutional failures. As a result, they have a construct in which they have no responsibility to identify and achieve optimal trade-offs that would involve explicit welfare comparisons and value judgments. Defining and assuming away the possible need for policy trade-offs does allow for neat, tractable analytical models. However the resulting policy prescriptions have contributed to decidedly sub-optimal outcomes.