Research
Current Working Papers
1. The Chinese Silver Standard: Parity, Predictability, and (In)Stability, 1912–1934 and its online appendix with Huachen Li (February 2024).
This paper assesses the debate about the demise of the Chinese silver standard in the mid 1930s. One side argues the U.S. Silver Purchase Act of June 1934 drained China of silver, which caused deflation and economic crises. A related claim is the Chinese silver standard was intrinsically unstable. These hypotheses are evaluated by estimating Bayesian structural VARs with drifting parameters on China-U.K. and China-U.S. samples from April 1912 to September 1934. We find instability in the Chinese silver standard peaked during the NBER recession of January 1920–July 1921 and the Great Depression (August 1929–March 1933). Hence, neither the U.S. Silver Purchase Act of June 1934 nor a design flaw lead to the end of the Chinese silver standard.
2. Online Supplement to UK Inflation Dynamics since the Thirteenth Century, with Gregor W. Smith (May 2023). The published paper is available at the International Economic Review 64(4), 1595-1614.
3. Appendix to Measuring the Slowly Evolving Trend in US Inflation with Professional Forecasts, with Gregor W. Smith (April 2020). The published paper is available online at the Journal of Applied Econometrics 36(1), 1-17.
4. Appendix, Supplement (June 2020) and Appendix, Additional Results (April 2020) to Inflation and Professional Forecast Dynamics: An Evaluation of Stickiness, Persistence, and Volatility with Elmar Mertens. The published paper is available online at Quantitative Economics 11(4), 1485-1520.
5. Sticky Professional Forecasts and the Unobserved Components Model of US Inflation, with Gregor W. Smith (October 2016).
This paper represents a fork in the road for our paper, Measuring the Slowly Evolving Trend in US Inflation with Professional Forecasts, which appears below. Gregor and I report tests of a joint model of U.S. inflation and professional forecasts of U.S. inflation. The joint model consists of an unobserved components model of inflation that decomposes it into trend and gap and a model of sticky information (SI). We show that trend inflation can be identify using inflation predictions at short horizon made by the average member of the Survey of Professional Forecasters (SPF). Similarly, inflation forecasts generated by the SI model can be tied to average SPF inflation predictions. Inverting the joint model produces estimates of the innovations to trend and gap inflation. Our tests ask if these innovations are serially correlated or persistent. The tests are robust to forecast horizon, stochastic volatility in trend and gap inflation, and place no restrictions on the covariance matrix of innovations to trend, gap, and/or SI inflation. On samples of GNP/GDP deflator inflation (CPI inflation) and the associated SPF predictions running from 1968Q4 (1981Q3) to 2016Q2, the results indicate U.S. inflation is sticky. However, the estimated innovations to trend and gap inflation are never jointly unpredictable. Changes to trend inflation exhibit statistically significant persistence. Our results suggest either the trend-cycle model of inflation is misspecified, the SI model is, or both.
6. Bringing Financial Stability into Monetary Policy, with Eric M. Leeper (July 2015). Revise and resubmit at the International Journal of Central Banking.
This paper arms central bank policy makers with ways to think about interactions between financial stability and monetary policy. We frame the issue of whether to integrate financial stability into monetary policy operating rules by appealing to the observation that in actual economies financial markets are incomplete. Incomplete markets create financial market frictions that prevent economic agents from perfectly sharing risk; in the absence of frictions, financial (in)stability would be of no concern. Overcoming these frictions to improve risk sharing across economic agents is, in our view, the intent of policies geared toward ensuring financial stability. There are many definitions of financial stability. Although the definitions share the notion that financial stability becomes an issue for policy makers when a breakdown in risk-sharing arrangements in financial markets has a negative effect on real economic activity, we give several examples that show this notion is too general for thinking about the role monetary policy might have in smoothing shocks to financial stability. Examples include statistical models that seek to separate “good” from “bad” changes in private-sector debt aggregates, new Keynesian policy prescriptions grounded in neo-Wicksellian natural rate rules, and a historical episode involving the 1920s Federal Reserve. These examples raise a cautionary flag for policy attempts to control the growth and the composition of debt that financial markets produce. We conclude with some advice for revising central banks’ Monetary Policy Reports.
Slides for Invited Lectures
1. Semi-Plenary Talk "Realized and SPF Inflation Dynamics: Methods and Applications" at the 2019 North American Summer Meeting of the Econometric Society (at the University of Washington, Seattle, WA), 28 June 2019. The attached slides are a revised version of the talk that markets papers by Elmar and me (see above) and Gregor Smith and me (see above). These projects begin from the premise that (a) inflation expectations express private agents’ beliefs about the underlying factors driving observed inflation dynamics and (b) these beliefs can be recovered from nonlinear state space models (SSMs) estimated on realized inflation and the average inflation predictions of the Survey of Professional Forecasters (SPF). Motivation is that professional inflation inflation predictions are often superior to model-based inflation forecasts. Professional inflation inflation predictions also contain information about the credibility of monetary policy in the sense that longer-horizon average SPF inflation predictions have often been less than 20 basis points from the FOMC’s 2% inflation target since 2012Q1. Next, the slides sketch statistical and economic models that combine realized inflation and average inflation predictions of the SPF developed in these papers. The models are internally consistent in that the statistical and economic models yield term structures of inflation expectations and forecasts that impose cross-equation restrictions on the state space models. This is followed by an outline of the sequential Monte Carlo methods that Elmar and I and Gregor and I employ to estimate the states and parameters of the nonlinear state space models. The slides finish with several results reported in the paper by Elmar and me.
2. Lecture on "Central Banker’s Modeling Toolbox: One-for-All or All-for-One?" at the Institute for Monetary and Economic Studies of the Bank of Japan, 12 October 2017. The IMES-Bank of Japan invited me to give this lecture, which is an extended version of the presentation I gave at the Bank of Canada in November 2016 (see below). The lecture is framed by the Folk Theorem that "All models are false." The question becomes how economists at central banks can use falsifiable models to advise policy makers. The first task is to review several approaches economists have to evaluate alternative policies conditional on “the models on the table.” A goal is to motivate discussion using a diverse set of models at central banks to evaluate monetary policy because a corollary to the Folk Theorem is "It takes a model to beat a model." This part of the lecture is also aimed at convincing economists to consider the ways in which policy makers engage with models. Next, there is a discussion of times series models with time-varying parameters(TVPs). Using models with TVPs gives economists a chance to analyze alternative policies that have the potential to alter the expectations of private agents. The lecture concludes with a survey of the current state of new Keynesian DSGE models used at central banks.
3. Lecture on "Bringing Financial Stability into Monetary Policy" at the Institute for Monetary and Economic Studies of the Bank of Japan, 13 October 2017. The IMES-Bank of Japan invited me to give this lecture, which is an updated version of the paper by Eric Leeper and me (see above). The lecture begins from the premise that in models with a complete set of Arrow-Debreu contingent claims markets bankruptcy and default are states of the world. However, agents fully insure against these events because the AD securities are a means to completely diversify risk. Since actual modern economies always face an incomplete set of AD markets, the goals of financial stability policies are to improve existing risk sharing arrangements and to smooth shocks that prevent or inhibit risk sharing. From this point, the lecture discusses problems of defining and measuring financial trends and financial cycles. One question is whether the tools modern students of the business cycle employ can be used to define and measure financial trends and financial cycles without theory and models. A related question is how to integrate financial trends and cycle into a model that decomposes macro aggregates into growth trends and business cycles. The rest of the lecture is dedicated to a (brief) review of the several theories and models financial and macro economists have developed to study financial frictions in general equilibrium. The lecture argues that these theories and models have difficulties explaining insolvency shocks as the source and cause of financial crises.
Slides for
Conference Presentations and Discussions
1. Discussion of the prospectus for inflation: “Is Inflation back with Vengeance?” at the Seventh Workshop on Prices and Inflation held virtually by the Centre for Prices and Inflation at Cardiff University and the U.K. Office of National Statistics on 5 July 2021. The organizers invited me to be part of a round table discussion that concluded the workshop. My discussion consisted only of two charts, which are the (same) linked slides. Please, note the slides have been updated as of 02 December 2023 to reflect (i) predictions by the (average of the) Survey of Professional Forecasters (SPF) through 2023Q4, (ii) BEA releases through 2023M10 and 2023Q3, and (iii) the NBER Business Cycle Dating Committee announcement that they “determined that a trough in monthly economic activity occurred in the US economy in April 2020.” (19 July 2021)
The top panel of the first figure plots monthly (year over year) headline chain-weighted Personal Consumption Expenditure (PCE) deflator inflation, PCE deflator inflation of services, PCE deflator inflation of nondurable goods, and PCE deflator inflation of durable goods along with the contributions of services, non-durable goods, and durable goods inflation to headline PCE deflator inflation from 1960M01 to 2023M10. The arithmetic of PCE deflator inflation shows that a switch from deflation to inflation in durable goods prices in the autumn of 2020 along with substantial increases in inflation of services and non-durable goods during the spring of 2021 contribute to higher inflation in the U.S. price index, the headline PCE deflator, that the FOMC has announced “is most consistent over the longer run with the Federal Reserve’s statutory mandate.” (27 August 2020 and reiterated 25 January 2022) Whether year over year realized headline PCE deflator inflation at 2.96% in 2023M10 after peaking at 6.88% in 2022M06 signals a continuing unwinding of U.S. inflation dynamics, is a question to be answered during 2024. The current almost year over year 3.0% growth in the PCE aggregate price deflator represents almost a 20 basis point decrease from a previous trough of 3.15% in 2023M06. Rising year over year realized headline PCE deflator inflation is attributed to year over year PCE-services deflator inflation dropping by 25 basis points from 4.60% in 2023M09 to 4.35% in 2023M10, which is the smallest year over year growth in this measure of prices since 2021M10, year over year PCE-nondurable goods deflator inflation rising from -0.29% in 2023M06 to 2.68% in 2023M09 before falling to 1.56% in 2023M10, and the PCE-durable goods deflator exhibits deflation with year over year growth of -0.54% in 2023M06, -2.34% in 2022M09, and -2.22% in 2023M10. However, year over year PCE-services deflator inflation, PCE-nondurable goods deflator inflation, and PCE-durable goods deflator inflation all are off peaks of 5.84 in 2023M02, 12.28% in 2022M06, and 10.19% in 2022M02. Hence, disinflation continues to be story of the services and nondurable goods components of the headline PCE price deflator for the last 10 to 18 months. More striking is the deflation in year over year PCE-durable goods deflator from 2023M06 to 2023M10. The current PCE-durable goods deflator matches its level in 2021M11 and 2021M12. It will reach its previous trough that occurred in 2020M06 in about 24 months by continuing to deflate by at least 2.0% per month (at an annual rate).
All this makes clear the three components of the headline PCE- deflator have behaved differently during the last 10 to 18 months. Year over year PCE-services deflator inflation has declined from 5.10% in 2023M07, to 4.71% in 2023M08, 4.61% in 2023M09, and 4.35% in 2023M10, but year over year PCE-nondurable goods deflator inflation has increased from 0.24% in 2023M07 to 2.11% in 2023M08 to 2.68% in 2023M09 before falling to 1.56% in 2023M10. As has been noted in the previous paragraph, PCE-durable goods deflator inflation has been been less than zero since 2023M06. In summary, these components of PCE deflator inflation put together show its headline version fell close to 400 basis points from a peak of 6.87% in 2022M06 to 2.96% in 2023M10. The PCE-services deflator, PCE-nondurable goods deflator, and PCE-durable goods deflator are responsible for this decline in the aggregate price index the FOMC has proclaimed to be its indicator of the nominal state of the U.S. economy. However, the drop in this aggregate price level masks substantial relative price movements across the PCE-services, PCE-nondurable goods, and PCE-durable goods price deflators during the last year to year and a half. The relative prices indicate that the PCE-services price deflator the PCE-nondurable goods deflator has exhibited more volatility compared with the PCE-services price deflator, but the relative price of durable goods has fallen against these components of headline PCE price deflator.
The monetary policy quandary, perhaps, is that during the last 12 months year over year growth in these three components of the headline PCE deflator have diverged. Does this suggest movements in the FOMC’s preferred aggregate price index are dominated by fluctuations in relative prices and shocks to these prices rather than the nominal shocks of the late 1960s, 1970s, and from 2020 to mid 2022? Are disturbances in relative price a real factor driving the U.S. business cycle? Can monetary policy respond repeatedly to real factors and remain credible? Is this an impediment to households, workers, firms, and investors believing the FOMC has genuinely recommitted itself to its inflation target of 2%?
The bottom panel of the figure provides more visual evidence on which to assess these questions. The share or contributions of PCE deflator inflation of services, PCE deflator inflation of nondurable goods, and PCE deflator inflation of durable goods to headline PCE deflator inflation appears in the bottom panel of the first figure from 1960M02 to 2023M08. Year over year PCE-nondurable goods deflator inflation is responsible for 0.34 percentage points of the 2.96% year over year realized headline PCE deflator inflation in 2023M10, which is a drop of 107 basis points since the start of the year and an increase of 41 basis points since 2023M06. Similarly, -0.27 percentage points of headline PCE deflator inflation is associated with PCE-durable goods deflator inflation in 2023M10. Hence, there is no surprise that the services component of the headline PCE price deflator has been and continues to be the major contributor to its year over year inflation rate. In 2023M10, 2.86 percentage points in year over year headline PCE deflator inflation is attributed to its services component. This accounts for about 97% of year over year headline PCE deflator inflation in 2023M10.
These plots also make it clear that episodes of inflation in the U.S. occur when the shares of the three components of headline PCE deflator inflation are rising at the same time. The first episode begins with the 1973M11-1975M03 recession and ends around the double-dip recessions of 1980M01-1980M07 and 1981M07-1982M12. The only other period with similar dynamics in the shares of the components of headline PCE deflator inflation starts in the fall of 2020 and appears to be finished by 2022M06. During the last 12 months, growth of nondurable and durable goods prices have contributed less and less to inflation in the aggregate price index to the point that by June the latter component started to deflate. A question going forward is whether inflation in the services, nondurable goods, and durable goods components of the headline PCE price deflator are on paths that will persist in diverging. The answer to the question matters because relative price shocks are, as hinted in the previous paragraph, neither caused by nor susceptible to the inducements of monetary policy. At the end of the day, monetary policy may be able to control the aggregate price or inflation and not much else, but may have other (not intended) consequences.
The second figure of the slides displays quarterly headline chain-weighted PCE deflator inflation from 2007Q1 to 2023Q3 and the average SPF nowcast, 1-quarter ahead, and 4-quarter ahead predictions of this measure of inflation from 2007Q1 to 2023Q4. The plots in the top panel show that realized PCE deflator inflation exhibits the most variation followed by the SPF nowcast. The 1-quarter ahead average SPF prediction is smoother, but smoothest is the 4-quarter ahead SPF prediction. The latter observation is consistent with results reported in the papers with Elmar Mertens and Gregor W. Smith mentioned above. These papers give evidence that the average 4-quarter ahead SPF inflation prediction is a reasonable and efficient proxy for trend inflation. For example, the average 4-quarter ahead SPF inflation prediction has increased by almost 100 basis points from 1.52% in 2020Q3 to 2.58% in 2023Q2, but fell to 2.46% in 2023Q3 and 2.32 in 2023Q4. The most recent predictions are more than 50 basis points lower than the previous peak of 2.98% in 2022Q4. Also, note year over year headline PCE deflator inflation was 2.45% in 2023Q2 and is 2.78% in 2023Q3, which is a drop of about 130 basis points from its realization of 4.08% in 2023Q1.
The bottom panel of the figure contains plots of average SPF nowcast, 1-quarter ahead, and 4-quarter ahead predictions minus the target 2% inflation rate of the FOMC. (24 January 2012, reiterated 27 August 2020, and again 25 January 2022). There are also 35 basis point upper and lower bands in the figure. The upper and lower bands are grounded in the assumption that about a third of the standard deviation of headline chain-weighted PCE deflator inflation from 2007Q1 to 2023Q4, which is about 1.1, is attributed to measurement error. The medians (means) of the average SPF-nowcast and 1- and 4-quarter ahead inflation predictions net of the target 2% inflation rate were -33 (-49), -24 (22), and -6 (-5) basis points from 2007Q1 to 2019Q4 with standard deviations of 59, 21, and 11 basis points. Hence, the average SPF 4-quarter ahead inflation predictions net of 2% were much less than 35 basis points (plus or minus) from 2007Q1 to 2019Q4. This suggests the average member of the SPF took the FOMC at its word about its 2% target for headline PCE deflator inflation. However, the average SPF 4-quarter ahead predictions cross the threshold of the upper 35 basis point band in 2022Q1. Although the average SPF 4-quarter ahead inflation predictions net of 2% was about 10 basis points below the lower band in 2020Q4, since the release of the FOMC statement about its longer run goals on 27 August 2020 the SPF nowcast and 1-month ahead prediction net of 2% have been above the upper band. The 4-month ahead prediction only fell on top of the upper band in 2023Q4. An interesting question is whether the most recent average SPF 4-quarter ahead prediction suggests trend inflation is running between 2.0% to 2.5% in the U.S. going into the summer of 2023. If true, trend inflation is nearer the 2% target for headline PCE deflator inflation since the FOMC announced its “long-run horizon” adjustment to its 2% target rate of headline PCE deflator price inflation.
Otherwise, I hope the figures are self-explanatory. Although the slides are silent on whether a combination of (a) supply chain disruptions, weak sectoral and aggregate productivity, a change in the composition of consumption, growth in the monetary aggregate M2, unfunded federal fiscal deficits, and/or a surplus of aggregate demand are the underlying sources and causes of the current state of the U.S. economy and (b) declining persistence caused the burst of inflation that began in 2021 and underpin the current state of the underlying stochastic process generating inflation in the aggregate price index in the U.S.. The arithmetic of the first figure is clear inflation is the state of the world in which all prices are continuously growing at the same time. This figure also suggests the sources and causes of the current underlying inflation regime appear to share much in common with the events associated starting with the post-Volcker disinflation of the late 1980s and 1990s. The second figure begs the question of whether the state of the world that matters most for the average member of the SPF is the regime shaping the underlying stochastic process generating U.S. inflation instead of the level of the FOMC’s policy rate. No matter the impact of relative price shocks (i.e., supply chain disruptions, weak sectoral productivity, changes in the composition of consumption, and growth in M2 and/or federal deficits), the slides suggest an interesting hypothesis or two. The first hypothesis asks whether the FOMC announcement of its longer run goals on 27 August 2020 (restated 25 January 2022) is an example of what Eric M. Leeper and Tao Zha (“Modest policy interventions,” Journal of Monetary Economics 50, 1673–1700) call an immodest policy. Subsequent to the FOMC announcement of 27 August 2020, realized inflation and average SPF inflation predictions began rising in the autumn of 2020. Should an economist steeped in the new Keynesian school think to test whether price setters of all kinds took this FOMC announcement as permission to revise their expectations about inflation, which lead them to exploit their powers in markets defined by monopolistic competition? Nevertheless, remember (dynamic cross) correlation is not causation. The same economist might want to examine a second hypothesis. The hypothesis tests whether movements in underlying trend inflation are unrelated to changes in and shocks to relative prices. Third, is it possible in real time to call when a the monetary policy regime of a central bank has achieved credibility.
2. Discussion of "Welfare Effects of Tax Policy in Open Economies: Stabilization and Cooperation" by Jinill Kim and Sunghyun Kim in the session, Monetary Policy & Fiscal Policy when Interactions Matter, at the annual IJCB Annual Research Conference, The Interplay between Monetary Policy and Fiscal Policy, at the Czech National Bank, Prague, Czech Republic, 20 June 2017. (Unfortunately, I was unable to attend the conference.) My extended comments on the Kim and Kim paper are found here.
3. Discussion of "Endogenous Technology Adoption and R&D as Sources of Business Cycle Persistence" by Diego Anzoategui, Diego Comin, Mark Gertler, and Joseba Martinez in the AEA Session Slowdown Risk: The Quest for Sustainable Growth, Chicago, IL, 6 January 2017.
4. Presentation of "Central Banker's Modeling Toolbox: One-for-All or All-for-One?" in Session II of the Workshop on Central Bank Models (click on the link), at the Bank of Canada, Ottawa, Ontario, Canada, 17 November 2016. The Bank of Canada hosted a one day workshop for its staff and academics to discuss the current and potential future state of finance, macro, and forecast models at central banks. I was fortunate to be invited by the Bank of Canada to participate in a session under the title that my presentation uses. For the session, the Bank of Canada staff posed four questions. The questions are
a) How should central banks manage the trade‐off between internal consistency of models and their complexity?
b) In what situations is it desirable to use a single model for projection and policy analysis, and in what situations is it desirable to use multiple models? Why? Is it true for both monetary policy and financial stability risk assessment models?
c) What is the desired composition of the central bank modeling toolbox 5 to 10 years from now?
d) Is a single large‐scale projection model suitable for assessing financial vulnerabilities and analyzing risks around the macroeconomic outlook? Could a set of targeted minimalistic models provide better risk analysis?
My presentation, which responds to these questions, are my own views and do not represent the views of the Bank of Canada and/or its staff.
5. Discussion of "Real-Time Forecast Evaluation of DSGE Models with Stochastic Volatility" by Frank Diebold, Frank Schorfheide, and Minchul Shin at the Conference, New Developments in Measuring and Forecasting Financial Volatility, Duke University and the University of North Carolina, Chapel Hill, held in Durham, NC, 16-17 September 2016.
6. Discussion of "In Search of a Nominal Anchor: What Drives Inflation Expectations?" by Carlos Carvalho, Stefano Eusepi, Emanuel Moench, and Bruce Preston at the Workshop on Methods and Applications for DSGE Models, Federal Reserve Bank of Philadelphia, 16-17 October 2015.
7. Discussion of "A Survey-Based Term Structure of Inflation Expectations" by S. Boragan Aruoba at the 10th Conference on Real-Time Data Analysis, Methods and Applications, Federal Reserve Bank of Philadelphia, 10-11 October 2014.
8. Discussion of "Evaluating Conditional Forecasts from Vector Autoregressions" by Todd Clark and Michael McCracken at the CIREQ Econometrics Conference, Time Series and Financial Econometrics, Université de Montréal, Montréal, Québec, Canada 9-10 May 2014.
9. Discussion of "Signaling Effects of Monetary Policy" by Leonardo Melosi at the Spring 2012 Bundesbank/Philly Fed Conference, Monetary policy, Inflation, and International Linkages, Bundesbank's Training Centre, Eltville am Rhein, Germany 24-25 May 2012.
10. Discussion of "Modeling Monetary Policy" by Samuel Reynard and Andreas Schabert at the Conference on Models and Policies for Economies with Credit and Financial Instability, Federal Reserve Bank of Cleveland, 14-15 October 2009.