New paper on the consequences of policy delay

My recent paper with Mai Hakamada of the IMF on monetary policy in the face of inflation surges, "The consequences of falling behind the curve: Inflation shocks and policy delays under rational and behavioral expectations," has been issued as IMF Working Paper WP/24/42 and is available at https://www.imf.org/en/Publications/WP/Issues/2024/03/01/The-Consequences-of-Falling-Behind-the-Curve-Inflation-Shocks-and-Policy-Delays-Under-545507?cid=em-COM-123-47987

Abstract: Central banks in major industrialized economies were slow to react to the surge in inflation that began in early 2021. The proximate causes of this surge were the supply chain disruptions associated with the easing of COVID restrictions, fiscal policies designed to cushion the economic impact of COVID, and the impact on commodity prices and supply chains of the war in Ukraine. We investigate the consequences of policy delay in responding to inflation shocks. First, using a simple three-period model, we show how policy delay worsens inflation outcomes, but can mitigate or even reverse the output decline that occurs when policy responds without delay. Then, using a calibrated new Keynesian framework and two measures of loss that incorporate a “balancedapproach” to weigh inflation and the output gap, we find that loss is monotonically increasing in the length of the delay. Loss is reduced if policy, when it does react, is more aggressive. To investigate whether these results are sensitive to the assumption of rational expectations, we consider cognitive discounting as an alternative assumption about expectations. With cognitive discounting, forward guidance is less powerful and results in a reduction in the costs of delay. Under either assumption about expectations, the costs of a short delay can be eliminated by adopting a less inertial policy rule and a more aggressive response to inflation. JEL: E31, E51, E52, E58, E61, Keywords: monetary policy; inflation; policy delay; behavioral expectations; falling behind the curve

What the House of Lord’s Report on the Bank of England got wrong

 

Over the past two years, many economies with independent central banks and formal inflation targets, including the UK, the U.S., and the Euro Area, have experienced inflation rates well above their proclaimed targets. While inflation rates are now declining, none of these central banks, or their leaders, have been held accountable for letting inflation soar to heights not seen for over 40 years.

Over the past summer things did heat up for Bank of England Governor Andrew Bailey. After a string of Bank forecasts that underestimated inflation, and pressure from the House of Commons Treasury Committee to conduct a review of its forecasting record, the Bank has established a committee chaired by Ben Bernanke to evaluate its forecasting methods. In addition, in November the Economic Affairs Committee (CEA) of the House of Lords issued “Making an independent Bank of England work better,” a report on the performance of the operational independence that was granted by the Bank of England Act 1998.

There is much to praise in the Lord’s report. Unfortunately, the report was marred by its analysis of inflation forecasting and by its inconsistent views on operational independence.

The Lord’s report criticizes the Bank’s forecasting record, noting that it, like other major central banks, systematically underestimated the level and the persistence of recent inflation. This isn’t surprising; did anyone predict the pandemic or Russia’s invasion of Ukraine? What is surprising was the call in the Lord’s for the Bernanke-led review “…to explain the value of models that seldom predict anything other than a return to the two per cent target over their forecast horizon.” (Para. 134) Why surprising? Whatever model the Bank uses to generate forecasts, those forecasts depend, explicitly or implicitly, on an assumption about monetary policy. What would it mean then if the Bank published a forecast for inflation that did not show a return to target? If the Bank’s is credibly committed to its inflation target, all its medium-term forecasts should show a return to two percent inflation. The most recent remit letter from the Chancellor to the Bank Governor actually, states “…monetary policy remains vital in supporting businesses and households by ensuring inflation returns to target sustainably in the medium term.” (emphasis added). For the Bank to forecast otherwise would be a red flag under any system of accountability. A private forecast might show inflation not returning to target, but that expresses a lack of confidence in the Bank’s commitment to its primary objective. A firm that signs an important contract to deliver goods at a future date but then immediately turns around and says they forecast delivering only half the promised amount would certainly be punished by financial markets.

A second issue with the Lord’s report is it seemingly calls for undercutting the Bank’s operational independence. The Bank’s primary objective – the inflation target – is set by HM Treasury, but with operational independence, the Bank has the freedom to act as it sees best to achieve that target. However, the report acknowledges that achieving the target may require the Bank to adopt a policy stance “... that counteracts the economic impact of the Government’s fiscal policy. It is therefore imperative that the Bank’s activities and its remit are effectively scrutinised by, and its officials are held accountable to, Parliament.” (para. 161) But that is exactly what operational independence is meant to prevent. It is the purpose of central bank independence, not a flaw, as the report recognizes when it states that “It is the Government’s job to ensure they (fiscal policy and the remit) are consistent with each other, not the Bank’s.” (para. 46) 

If the Government adopts an expansionary fiscal stance that the Bank believes will lead inflation to rise above target, the Bank must raise interest rates to offset the fiscal expansion. If it can’t do that for fear of being scrutinized by the fiscal authorities, what is the meaning of operational independence? And how then can the Bank be held accountable based on whether it achieves the inflation target? 

The report contains very little discussion on how to hold the Bank accountable. The issue is not whether an independent central bank should be accountable. Of course it should be. The issue is how accountability should be enforced. The original Bank of England Act 1998 Act established that the Governor write a letter to the Chancellor explaining any deviation from the inflation target of one percentage point or more. Being forced to write a letter seems like a very weak disciplinary punishment for letting inflation average almost 6 percentage points above the target since February 2021. Would the CEO of a private company whose profit guidance was that far off for almost two years still have a job?

Who called it the 'Hahn problem'?

In chapter 2 of my book Monetary Theory and Policy (The MIT Press, 4^th ed., 2017), I called the problem of establishing a positive value for money the 'Hahn problem'. Hahn described this problem in “On Some Problems of Proving the existence of an equilibrium in a monetary economy,” published in The Theory of Interest Rates, F. H. Hahn and F. P. R. Brechling eds. London: Macmillan, 1965. Pp. 126- 135. In the 4th edition of Monetary Theory and Policy, the reference to the Hahn problem appears on p. 41, and I credited Truman Bewley with naming it in his paper “A difficulty with the optimum quantity of money”, Econometrica 1983, 51(5), 1485-1504. 

Recently, Pierrick Clerc of the HEC Liège School of Management (https://sites.google.com/site/pierrickclerc/) has pointed out to me that the term was used a decade earlier by Kevin Sontheimer in “The determination of money prices”, Journal of Money Credit and Banking, 1972, 4(3), 489-508. Sontheimer solves the Hahn problem by employing a model in which there are costs to transacting that take the form of foregone leisure. This means his model falls within the general class of shopping time models discussed in section 3.2.1 of chapter 3 of Monetary Theory and Policy.

I would like to thank Pierrick for sending me the Sontheimer paper.

Claudia Goldin and measuring the experience of women in the labor force

 

Congratulations to Claudia Goldin, the 2023 recipient of the Nobel Prize in Economics. Goldin was cited “for having advanced our understanding of women’s labour market outcomes.” The press release from the Royal Swedish Academy of Sciences highlighted that she “provided the first comprehensive account of women’s earnings and labour market participation through the centuries. Her research reveals the causes of change, as well as the main sources of the remaining gender gap.” Goldin’s work has shown how women’s historical contributions to the labor force frequently went unmeasured, illustrating how what is not counted often ends up not counting in policy discussion.

As a Lecturer at Auckland University in New Zealand in the late 1970s, I encountered an example of how official statistics could conceal women’s labor market experiences. At that time, unemployment was calculated as the number of workers who had officially registered for unemployment benefits. However, unemployment benefits were effectively limited to heads of households. Secondary workers were not eligible for unemployment benefits and therefore had no incentive to register if they lost their jobs. And any unemployed worker who failed to register disappeared from the measured labor force. This approach meant the true extent of unemployment was understated, and the understatement was likely to be largest among women as they were less likely to register because they were more likely to be classified as secondary workers and not qualify for unemployment benefits. 

I attempted to overcome these measurement issues to obtain a more inclusive estimate of New Zealand’s unemployment rate. The results were published in 1978 in the New Zealand Economic Papers (“Unemployment in New Zealand: An Errors in Variables Approach to Measuring the Number of Unemployed,” NZEP, 12:1, 13-48). Based on data from 1965-1976, my estimates suggested New Zealand’s official unemployment figures captured about half of total male unemployment but only a tenth that of females. I also found the female unemployment rate over the estimation period was roughly three times the rate for males. 

With the election of David Lange’s government in 1984, New Zealand entered a period of economic reform, not least of which was the Reserve Bank Act of 1989 that established formal inflation targeting. And in 1985, New Zealand introduced a household survey to measure labor participation and unemployment so that their statistics could be brought in line with international best practice, making it no longer necessary to rely on the admittingly crude approach I had to use in the late 1970s. 

From 1988 to 2001, the female unemployment rate was less than the rate for males. Since then, the reverse has been true, though in general, the unemployment rates for both groups have been similar. Importantly, these data are now subject to less systematic measurement bias and therefore provide more accurate, and useful, information on the experience of women in the New Zealand economy.

New version of paper on super-active fiscal policy

Roberto Billi and I have a new version of our paper "Seemingly Irresponsible but Welfare Improving Fiscal Policy at the Lower Bound: The Role of Expectations." The addition of "The Role of Expectations" to the title emphasizes that a key contribution of the paper is an evaluation of how cognitive discounting affects the performance of an active fiscal policy, passive monetary policy regime in the face of occasional periods at the zero lower bound. The new draft also provides a more integrated introduction and literature review, as well as a streamlined discussion of the basic intuition for the results. It is available at https://people.ucsc.edu/~walshc/MyPapers/fslb_231010.pdf or on Roberto's home page at http://www.rmbilli.com/.

Keywords: automatic stabilizers, cognitive discounting, fiscal and monetary interactions, government debt. JEL: E31, E52, E63.

Is it too soon to revisit the Fed’s 2020 monetary policy framework? No!

 

On August 27, 2020, the FOMC released a revised Statement on Longer-Run Goals and Monetary Policy Strategy. This document laid out the goals for monetary policy, articulated the policy framework, and was designed to serve as the foundation for the Committee's policy actions. The statement indicated that the FOMC intends to conduct a regular review of its monetary policy strategy, tools, and communication practices roughly every five years.

That means that the next review may take place in 2025, and, if the past review is a guide, the Fed will launch the review in early 2024 as it prepares the groundwork for (possibly) revising its policy framework. It is time to start that review now.

Outside the Fed, the review has already started. The Brookings Institution, on May 23, 2023, hosted a conference titled “The Fed: Lessons learned from the past three years.” Gauti Eggertsson and Don Kohn discussed how the FOMC’s 2020 policy framework may have contributed to its delay in responding to the surge in inflation experienced in 2021, while a panel consisting of Ben Bernanke, Olivier Blanchard, Rich Clarida, Kristen Forbes, and Ellen Meade discussed potential reforms of the FOMC’s policy framework.

With hindsight, the 2020 review resulted in a policy framework that was designed, quite clearly, to fight the last monetary policy war, the battle against an inflation rate that was systematically too low in an environment of low interest rates. It was ill-suited for the current environment of high inflation. The next policy framework needs to be robust, not tailored to one set of circumstances that could quickly change.

At its core, any monetary policy framework needs to ensure inflation remains low and stable, promote financial stability, and make certain the central bank has the flexibility to respond in the face of shocks to the macro economy. It should also facilitate the central bank’s ability to clearly communicate policy to the public.

The shift in 2020 from inflation targeting to a form of average inflation targeting (AIT) meant the Fed’s inflation goal became ill-defined and, therefore, less transparent. At the Brookings event, Kristen Forbes and Rich Clarida both noted that a target range for inflation should be given serious consideration. A range would be easy for the public to understand, while avoiding the specious appearance that the Fed can preciously control inflation.

A new statement on goals and strategy must also address the “maximum employment” part of the Fed’s Congressional mandate. This part of the dual mandate has always been harder to translate into a specific measurable objective. If the FOMC wishes to maintain its asymmetric “shortfalls of employment” language, it needs to explain more clearly the basis on which it will judge whether the economy is short of its maximum employment or not. Unemployment rates that are at historically low levels as seen recently do not require a monetary policy response when inflation forecasts remain consistent with inflation goals. However, the Fed’s poor ability to forecast inflation makes such a strategy problematic.

Besides reviewing its policy strategy, the FOMC should examine its tactics. In 2021 it argued the shocks to inflation were temporary, justifying its failure to react. However, as I discussed in "Implications of a Changing Economic Structure for the Strategy of Monetary Policy", a paper presented at the Federal Reserve Bank of Kansas City’s 2003 Jackson Hole Symposium, better outcomes can be achieved by acting as if inflation shocks will not be temporary. When faced with uncertainty about the persistence of exogenous shocks, it is better to over-estimate the shock’s persistence, not underestimate if as the FOMC did, as it was consistently surprised when shocks failed to fade away. In that same paper, I showed that a central bank seeking a policy that is robust to uncertainty about shock persistence should act as if the shock will be more persistent than it really believes. That is, even if the policymaker’s forecast is that an inflation shock will quickly dissipate, policy should be designed as if the shock will persist.

Determining exactly what policy framework should replace the 2020 one requires care review. It is time to start undertaking that review.

Robert Lucas and the rational expectations revolution

 

Few economists over the past 50 years had as great an impact on macroeconomics as Bob Lucas, who died Monday May 15, 2023 at the age of 85. When Lucas received the 1995 Nobel Prize in Economics, the Prize committee cited him “for having developed and applied the hypothesis of rational expectations, and thereby having transformed macroeconomic analysis and deepened our understanding of economic policy.”

Lucas’s 1972 paper, “Expectations and the neutrality of money,” (Journal of Economic Theory 4(2), 103-124) set out a coherent theory reconciling the long-run neutrality of money with the short-run non-neutrality of money. He did so by carefully modeling an informational friction that generated expectational errors in the short-run, errors that, because individual firms and households formed expectations rationally, could not persist. In his model, only monetary policy surprises mattered for the real economy. This result had profound implications for monetary policy. Lucas and Sargent summarized these implications in 1978, noting that policy had to be systematic, stating that “… [equilibrium methods] will focus attention on the need to think of policy as the choice of stable rules of the game, well understood by economic agents. Only in such a setting will economic theory help predict the actions agents will choose to take.” Yet policy must also be unpredictable: “the government countercyclical policy must itself be unforeseeable by private agents...while at the same time be systematically related to the state of the economy. Effectiveness, then, rests on the inability of private agents to recognize systematic patterns in monetary and fiscal policy (Lucas and Sargent 1978).

When expectations are formed rationally, having policy be systematic while also being unpredictable is a contradiction. Whatever systematic rule the central bank chose to follow would come to be understood by the public and would then make no difference for the real economy; it would matter for inflation, but not for output or unemployment. This conclusion became known as the policy irrelevance hypothesis.

Lucas offered empirical evidence for his theory in “Some International Evidence on Output-Inflation Tradeoffs.” (American Economic Review, 1973, 63 (3): 326–34). There he showed that evidence on the slope of the output-inflation tradeoff from a cross-section of countries followed “directly from the view that inflation stimulates real output if, and only if, it succeeds in "fooling" suppliers of labor and goods into thinking relative prices are moving in their favor.”

The development of the new Keynesian approach during the 1990s shifted the focus of the non-neutrality of monetary policy from informational frictions to price and wage rigidities. Today’s models are firmly based on rational expectations, relying on nominal rigidities in the form of sticky prices and wages as the rationale for the potentially significant but ultimately temporary effects of monetary policy on the real economy.

Importantly, the emphasis today is that policy must be be systematic and predictable if it is to be effective. As described by Woodford in Interest and Prices: Foundations of a Theory of Monetary Policy (2003, Princeton, NJ: Princeton University Press), “...the central bank’s stabilization goals can be most effectively achieved only to the extent that the central bank not only acts appropriately, but is also understood by the private sector to predictably act in a certain way. The ability to successfully steer private-sector expectations is favored by a decision procedure that is based on a rule, since in this case the systematic character of the central bank’s actions can be most easily made apparent to the public.” (p. 465, emphasis in original).

While Lucas’s model of monetary non-neutrality has largely been replaced, his impact on current policy discussions is still strong. Discussions of monetary policy, whether in theoretical work or in policy practice, continue to stress the importance of policy credibility and of behaving in a systematic fashion. When central bankers emphasize the important role of inflation expectations, the need to anchor expectations, and that a credible commitment to lower inflation will lessen the unemployment costs of anti-inflation policies, they are channeling the fundamental insights of Bob Lucas.

Wages and inflation

Will faster wage growth lead to higher inflation? Discussions of wage and price spirals are in the news, but by comparing the growth rates of wages and prices, one can lose sight of what is going on with the levels. Real wages – the level of wages relative to the price level – are what matter for those receiving labor income. And what has happened to real wages during the past two years of above target inflation depends very much on which price index one uses to deflate nominal wages.

The Fed’s preferred index for measuring inflation is the personal consumption price index less food and energy prices, called core PCE). The PCE index does a better job of capturing how household substitute away from goods whose prices have risen more than average and away from goods whose prices have risen less than average. The ability to substitute across goods helps to mitigate the effect of inflation on households’ budgets. Eliminating the prices of food and energy, which tend to be volatile, can also give a better picture of future inflation. Using core PCE to deflate the BLS’s series on average hourly earnings yields an estimate of the real wage that is shown in blue in the figure. All data are monthly, seasonally adjusted and obtained from the St. Louis Fed’s FRED database; series are normalized to equal 100 in January 2020. The real wage series based on the core PCE price index suggests wages have kept up with inflation. Real wages spiked during the COVID recession and then dropped in its aftermath, but the measure has remained essentially flat since June 2020.

However, core PCE is not the only way to measure inflation. A more common measure of inflation, so common it is sometimes referred to as headline inflation, is based on the Consumer Price Index (CPI). This index does not do as good a job as the PCE does in incorporating how households shift their spending patterns in response to changing prices. Substituting to cheaper alternatives may cushion the impact of inflation, and this margin of adjustment is better captured by the chain-index PCE inflation measure, but the fact that a household needs to make such a substitution may make the price rise more salient in the household’s assessment of their cost of living. If so, CPI inflation might better capture how households assess the impact of inflation on their living standards. The red line in the figure shows what has happened to the real wage when the CPI is used instead of core PCE. The CPI-based real wage shows a steady decline between June 2020 and June 2022. Since August 2022, the CPI measure of real wages has remained essentially flat at a level approximately 4% below its June 2020 level.

Which measure of real wages might be most relevant for foretelling future wage growth? According to a Gallup Poll (April 6, 2023), inflation remains the top economic concern of Americans in 2023 as it was in 2022, with the fraction of respondents citing inflation as something they worry about “a great deal” coming in at 59% in 2022 and 61% in 2023. The number citing unemployment as something they worry about a great deal was just 32% in both 2022 and 2023.

It seems hard to reconcile how many American are worried about inflation with the stability of the real wage based on the Fed’s preferred price index. The decline in the CPI-based real wage seems much more consistent with the Gallup data. If this is the case, greater pressure for wage growth that is sufficient to boost real wages seems likely.