Yellen's IMF Speech on Financial StabilityThe topic of Yellen’s speech, financial stability, is particularly appropriate given the current critique that highly accommodative policies among the big four central banks are resulting in overvaluations in various markets (eg, housing, bonds, credit, equities) and prompting households and firms to take undue risks.
Just this weekend, at the start of the BIS AGM, General Manager Jaime Caruana summarized this critique succinctly. Speaking of the various effects of “extraordinarily accommodative” monetary policy, he noted, “There is a threat to financial stability too, as ultra-low interest rates promote debt accumulation and risk-taking.” And he concluded his comments on monetary policy by suggesting, “After years of easy money, we need to pay more attention to the risks of normalising too late.”
Yellen’s recent comments suggest she’s currently more sanguine about these risks. For example, at the June 18 FOMC press conference, Greg Ip asked Yellen specifically, “Will financial stability considerations play a role in when and how fast the Committee normalizes interest rates?” Yellen’s response was fairly direct. “I don’t see them shaping monetary policy in an important way right now. I don’t see a broad-based increase in leverage, rapid increase in credit growth or maturity transformation, the kinds of broad trends that would suggest to me that the level of financial stability risks has risen above a moderate level.”
However, the larger discussion among central bankers is about the ways in which financial stability should influence their overall policy frameworks. For example, simple rule-based approaches, such as the Taylor rule, don’t address the issue of financial stability. And yet many central bankers believe there is a political and social imperative post crisis to include financial stability in their policy frameworks. The challenge for each central bank then is to develop a framework that addresses financial stability in a way that reflects its particular mandate.
The most active line of research on this issue within the Fed involves the use of a loss function that penalizes the policy maker for deviations from his mandated objectives. In the quadratic version of this loss function (discussed most frequently at the Fed), the penalty increases as the square of the deviation. For example, if the inflation objective is 2% and the actual inflation rate is 3.5%, the penalty is considered to be 2.25 (ie, [3.5-2]^2 ). The strategy of the policy maker then is to minimize the sum of the expected deviation from target and the expected value of the squared deviation from target (ie, the variance). In the case of the Fed, there would be two targets: inflation, and employment.
The relevant aspect of this approach is that it presents the policy maker with a trade-off between the mean and the variance of the deviations from target. As a result, the policy maker may intentionally follow a policy in which inflation and/or unemployment are expected to deviate from their targeted values if the policy results in a reduction in the variances of these deviations from their targets. Or as Kocherlakota put it in a speech in April of last year, “…the Committee has to weigh the certainty of a costly deviation from its dual mandate objectives against the benefit of reducing the probability of an even larger deviation from those objectives.”
There are quite a number of potential problems in adopting this approach. For example, the Fed would need to have models for inflation and unemployment that specify their variances as well as their expected values. In a speech in March of this year (ie, while still a Board member), Jeremy Stein suggested that bond term premiums and credit spreads would be useful variables in this context. In particular, he suggested, “…monetary policy should be less accommodative--by which I mean that it should be willing to tolerate a larger forecast shortfall of the path of the unemployment rate from its full-employment level--when estimates of risk premiums in the bond market are abnormally low.”
Stein’s comment highlights not only the modelling challenges inherent in the loss function approach but also the political difficulties the Fed is likely to encounter were it to adopt this framework. For example, one can imagine the heated comments Yellen would receive from members of Congress during her semi-annual testimony if she acknowledged that the Fed was intentionally sacrificing the jobs of their constituents because the yield curve was too flat.
An additional economic and political challenge would be to specify an appropriate employment target, which presumably would need to vary over time. And for a Fed so clearly committed to transparency, attempts to communicate policy formulated with a quadratic loss function would pose a considerable challenge.
Given Yellen’s recent comments, I don’t expect her to announce a major shift in the Fed’s framework in today’s IMF speech. But I do believe it’s possible that she’ll provide an update on Fed research into this issue, as part of an evolution in the Fed policy framework, similar to the evolution in the Fed’s specification of a 2% inflation objective under Bernanke.
As for the immediate implications, I expect Yellen will reiterate that the observed degrees of indebtedness and leverage in the economy don’t point to the sorts of financial imbalances that would pose a risk to financial stability. Note that in her recent comments she’s avoided using asset prices per se as a metric for assessing financial stability, preferring instead to focus on the exposures within the household, financial, and corporate sectors.
Thursday's Nonfarm Payrolls ReportIn this context, it’s also worth considering the immediate implications of this week’s payrolls report on Fed policy. The median estimate of economists surveyed by Bloomberg is currently 213 K, with the unemployment rate remaining at 6.3%. Figures in line with these expectations would be consistent with a continuing decline in the available slack in the labor markets.
But the problem facing the FOMC is not to estimate the pace with which resource slack is declining. Rather, it’s to estimate the point at which the slack will be depleted, at which time increasing labor costs would be expected to exert upward pressure on the inflation rate. And to estimate the level of labor slack (as opposed to its change), one needs to take a view on the effect of quite a number of factors, such as the pace of retirements, hysteresis, and changes to various government programs, such as unemployment, disability, SNAP, and health care.
Given the difficulty of estimating labor slack directly, the Fed also finds it useful to monitor wage developments, given the reasonable expectation that unit labor costs will begin to increase as slack in the labor market is eliminated. For example, during her testimony before the Joint Economic Committee on May 7, Yellen noted, “most measures of labor compensation have been rising slowly--another signal that a substantial amount of slack remains in the labor market.”
Yellen also referenced the unemployment rate and labor costs during the press conference following the June 18 FOMC meeting. On the unemployment rate, she offered, “…many of my colleagues and I would see a portion of the decline in the unemployment rate as perhaps not representing a diminution of slack in the labor market.” And she later noted, “My own expectation is that as the labor market begins to tighten, we will see wage growth pick up … so households are getting a real increase in their take-home pay. She continued, “Within limits, it’s not a threat to inflation…” She then noted, “If we were to fail to see that, frankly, I would worry about downside risk to consumer spending.”
Financial Stability and Employment in the Context of Monetary PolicyAs the unemployment rate continues to decline, and as real and financial assets continue to richen, I expect the calls for the FOMC to hike rates will grow louder, including calls from members of the Committee. But I also expect that Yellen will continue to argue that unit labor costs are more useful than the unemployment rate in gauging slack in the labor market and that exposures in the household, financial, and corporate sectors are more important than asset prices in assessing risks to financial stability.
I expect these developments will occur in the context of a continuing trend in the asset markets in which the capital available for investment exceeds the demand for capital to invest, resulting in a general reduction in expected returns across asset markets, consistent with higher equity prices, tighter credit spreads, higher house prices, and lower real yields. But as the inflation hawks become increasingly vocal in H2, I expect we’ll see some upward pressure on the 1Y-5Y segment of the yield curve and upward pressure on longer-dated breakeven inflation rates. I also expect we’ll see greater volatility along the curve in general, particularly in the 1Y-5Y segment.