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.
The Federal Reserve Bank of Boston recently published a paper titled: “Reasonable People Could Disagree: Optimism and Pessimism About the US Housing Market Before the Crash.” The paper proposes two interrelated standards by which the Fed’s role in the run-up to the recent crisis can be judged. The authors “conclude by arguing that economic theory provides little guidance as to what should be the correct level of asset prices – including housing prices. Thus, while optimistic forecasts held by many market participants in 2005 turned out to be incorrect, they were not ex ante unreasonable.” The implied pass issued to the Fed for having missed the housing bubble elicited much anger and consternation from bloggers and others. Much, but not all of the criticism was directed at the failure of the Fed to recognize the bubble for what it was and for the paper’s denial of the Fed’s responsibility for its contribution to the crisis. The relatively narrow focus of the responses is unfortunate. A brief analysis of the lines of argument advanced by the authors reveals weaknesses in the argument and a decided shortcoming in the Fed’s approach to policy formulation prior to the crisis.
As part of the first standard (consensus of economists) by which the Fed’s role can be judged, the authors argued that economists were divided over the existence of a housing bubble:
i) some economists believed and argued that a bubble existed – the pessimists;
ii) a larger group of economists did not believe a bubble existed – the optimists;
iii) most economists were agnostic on whether or not a bubble existed.
Based on the divisions among economists and the high concentration of agnostics, the authors drew their conclusion: given the absence of a consensus it was not unreasonable for the Fed to adhere to the optimistic view — there is no bubble. This line of reasoning assumes that it is appropriate for a policymaker at the Fed to view and react as economists would, however nothing could be farther from the truth.
Economists have every right to hold any opinion about markets that they choose, they have every right to express it, and they have every right to be agnostic about any or every policy or issue. However, as economists they have no responsibility except to themselves and their careers. In fact, the paper suggested that most economists decided not to take a position on the existence of the housing bubble because of reputational risk, i.e., they were afraid to be wrong.
The Fed (and economic policymakers in general) is in a very different position. It does not have a right to have an opinion or not; it has a responsibility: promoting full employment and price stability. Asset prices are drivers of economic activity — business investment, residential investment, and consumption as well as determining the value of collateral that supports the financial system. Consequently, the Fed cannot fulfill its mandate by being agnostic about the possible existence of macro-economically important asset price bubbles or other economic or financial imbalances. In order to meet its responsibility, the Fed must incorporate the possible existence of an asset bubble or imbalance in to the policy formulation process. The fact that economists had not reached a consensus on the existence of a housing price bubble is irrelevant to the assessment of the Fed’s role prior to the crisis.
In the second proposed economic policy evaluation standard, the authors argued that economics did not provide an analytically based means of identifying asset price bubbles. (The consensus of economists’ standard discussed above may be viewed as a corollary of this standard.) The authors argued that economic theory of asset prices is insufficiently well developed to distinguish “bubble” prices from equilibrium or near-equilibrium prices. The authors then go on to argue that this represents a theoretically based justification for both economists to have been agnostic and the Fed to have been acting reasonably when it failed to see and respond the housing bubble. (Note: At the time that the Boston paper was published, Fed officials were actively arguing that there was no bubble in the prices of Treasuries.)
While the Boston Fed paper is correct in that asset price theory does not provide a means to identify bubbles with any degree of certainty, it also ignores an equally important point. Asset price theory could not rule out the existence of the bubble with certainty either. Given the absence of certainty, the process of weighing the merits of alternative policies and choosing between them should reflect the examination and weighting of all possible outcomes across all the possible policy choices and possible states of the world and not simply the outcomes associated with the assumed to be most likely current states of the world (e.g. there is no housing price bubble).
However, history suggests that the Fed assigned a higher probability to the non-existence of the bubble than it did to the existence of the bubble and then continued to set policy and otherwise act as if it knew with certainty that no bubble existed. This is reflected in policy stances (monetary and regulatory), the language in speeches given by members of the FOMC and is also consistent with the argument presented in the Boston Fed paper. If so (and the evidence indicates it was), then the Fed shirked its responsibility. In particular, evidence is consistent with the position that the Fed failed to include in its policy calculus the answer to a question that it should have, but presumably never got around to asking: What would be the implications for the housing market, the real economy, financial markets and institutions, if the Fed continues to set policies assuming that there is no bubble in real estate prices when in fact there is an unsustainable price bubble?
If the Fed had employed its tools and powers as a regulator and supervisor to research the answer to some a few aspects of that question, it would have discovered that many important players, including major financial institutions that it regulated, were not just “optimists”, but were increasingly highly leveraged, maturity-mismatched, bet-the-firm “optimists”. (Note: the Fed has argued that information it gleans as a regulator is important in the formulation of monetary policy. In fact, this is the principle argument it has used for retaining a regulatory role.) It also would have been able to make educated guesses about the risks inherent in some bank counterparty positions.
Combining the information available to it as a regulator with information from other regulators and publicly available information on the size of down payments, negative amortization loans, teasers rates etc., the Fed would then have been able to assess the risk it was running when it continued to set policies assuming that there was no bubble in real estate prices. It would have been in a position to foresee some of the implication of a future end to house price appreciation for mortgage defaults, losses given default, the mortgage market, the housing market, the real economy, as well as on financial instruments, markets and institutions. It would have been able to recognize the possibility of a self-reinforcing downward pressure on the prices of housing and housing-related assets that would have manifested themselves if price appreciation faltered for any reason.
If the Fed had performed the analysis, it would have discovered at least some of the growing fault lines running through the housing market and the financial system. Enough relevant information was available to the Fed to set off policy alarm bells. If the Fed had done this research, it would not have been surprised by the crisis. Had the Fed performed the analysis, it is likely that monetary and or regulatory policies would have been different than they were.
In short, history supports the argument that the Fed decided that the probability of the existence of a real estate price bubble was less than the probability of no price bubble and that it never assessed the costs to society that would arise if it continued to set policy predicated on sustainable real estate prices when in fact there was a price bubble. As a result, it incorrectly estimated the expected pay-off to maintaining policy stances that assumed no real estate bubble existed as well as underestimating the risks associated with those policy stances.
Back to the Boston Fed paper, the absence of an analytic “solution” to the question “Is there an asset price bubble?” is largely irrelevant in assessing the Fed’s role prior to the crisis. The Fed failed to ask the correct questions. There were other perspectives and tools available. The Fed did not exhaust them.
Policymakers must be prepared to act when formal economics gives little or no guidance. To do that, policymakers must be willing to make decisions despite the limitations of economics, to use types of information that economists do not use or do not have access to, have the courage to act despite incomplete information and in the presence of risk and Knightian certainty. Policymakers must not assume certainty when there is none. Agnosticism and policymaking do not mix. Policymakers do not have the option of sitting on their hands, hoping for the best while trying to avoiding reputational or political risk. Policymakers cannot escape responsibility for economic underperformance, simply because economics doesn’t provide a simple unambiguous policy rule.