-
Working Paper
01.04.2024 |
WP 24-02
For decades, the Federal Reserve Bank of Cleveland (FRBC) has produced median and trimmed-mean consumer price index (CPI) measures. These have proven useful in various contexts, such as forecasting and understanding post-COVID inflation dynamics. Revisions to the FRBC methodology have historically involved increasing the level of disaggregation in the CPI components, which has improved accuracy. Thus, it may seem logical that further disaggregation would continue to enhance its accuracy. However, we theoretically demonstrate that this may not necessarily be the case. We then explore the empirical impact of further disaggregation along two dimensions: shelter and non-shelter components. We find that significantly increasing the disaggregation in the shelter indexes, when combined with only a slight increase in non-shelter disaggregation, improves the ability of the median and trimmed-mean CPI to track the medium-term trend in CPI inflation and marginally increases predictive power over future movements in CPI inflation. Finally, we examine the practical implications of our preferred degree of disaggregation. Our preferred measure of the median CPI suggests that trend inflation was lower pre-pandemic, while both our preferred median and trimmed-mean measures suggest a faster acceleration in trend inflation in 2021. We also find that higher disaggregation marginally weakens the Phillips curve relationship between median CPI inflation and the unemployment gap, though it remains statistically significant.
-
Working Paper
11.06.2023 |
WP 20-22R2
Accurate rent measurement is essential for constructing a consumer price index (CPI) and for measuring household welfare. Late payment fees and nonpayment of rent are common components of rental expenditures and thus belong in CPIs. Late payment fees are often excluded; we offer a novel critique. In the US CPI, nonpayment is ostensibly included, but, we show, severely undermeasured. Moreover, the manner of its inclusion renders the CPI extremely sensitive to nonpayment variations; we show how to fix this. Nonpayment undermeasurement suggests at least a +1 ppt overestimate in 2020 CPI shelter inflation. Timely nonpayment and late fee measurement is challenging; we offer a practical solution.
-
Working Paper
09.28.2023 |
WP 22-38R
Rent measurement determines 32 percent of the CPI. Accurate rent measurement is therefore essential for accurate inflation measurement, but the CPI rent index often differs from alternative measures of rent inflation. Using repeat-rent inflation measures created from CPI microdata, we show that this discrepancy is largely explained by differences in rent growth for new tenants relative to all tenants. New-tenant rent inflation provides information about future all-tenant rent inflation, but the use of new-tenant rents is contraindicated in a cost-of-living index such as the CPI. Nevertheless, policymakers should integrate new-tenant inflation into inflation forecasts and monetary policy decisions.
-
Working Paper
06.20.2023 |
WP 23-06R
We implement a novel nonlinear structural model featuring an empirically-successful frequency-dependent and asymmetric Phillips curve; unemployment frequency components interact with three components of core PCE – core goods, housing, and core services ex-housing – and a variable capturing supply shocks. Forecast tests verify model’s accuracy in its unemployment-inflation tradeoffs, crucial for monetary policy. Using this model, we assess the plausibility of the December 2022 Summary of Economic Projections (SEP). By 2025Q4, the SEP projects 2.1 percent inflation; however, conditional on the SEP unemployment path, we project inflation of 2.9 percent. A fairly deep recession delivers the SEP inflation path, but a simple welfare analysis rejects this outcome.
-
Working Paper
05.30.2023 |
WP 22-23R
Theory and extant empirical evidence suggest that the cross-sectional asymmetry across disaggregated price indexes might be useful in forecasting aggregate inflation. Trimmed-mean inflation estimators have been shown to be useful devices for forecasting headline PCE inflation. But is this because they signal the underlying trend or because they implicitly signal asymmetry in the underlying distribution? We address this question by augmenting a "hard" to beat benchmark headline PCE inflation forecasting model with robust trimmed-mean inflation measures and robust measures of the cross-sectional skewness, both computed using the 180+ components of the PCE price index. Our results indicate significant gains in the point and density accuracy of PCE inflation forecasts over medium- and longer-term horizons, up through and including the COVID-19 pandemic. Improvements in accuracy stem mainly from the trend information implicit in trimmed-mean estimators, but skewness information is also useful. An examination of goods and services PCE inflation provides similar inference.
-
Working Paper
02.14.2023 |
WP 19-09R2
We establish that the Phillips curve is persistence-dependent: inflation responds differently to persistent versus moderately persistent (or versus transient) fluctuations in the unemployment rate gap. This persistence-dependent relationship appears to align with business-cycle stages and is thus consistent with existing theory. Previous work fails to model this dependence, thereby finding numerous "inflation puzzles" – e.g., missing inflation/disinflation – noted in the literature. Our specification eliminates these puzzles; for example, the Phillips curve has not weakened, nor was inflation "stubbornly low" in 2019. The model's coefficients are stable, and it provides accurate conditional recursive forecasts through the Great Recession. There are important monetary policy implications.
-
Working Paper
01.13.2023 |
WP 23-06
In the December 2022 Summary of Economic Projections (SEP), the median projection for four-quarter core PCE inflation in the fourth quarter of 2025 is 2.1 percent. This same SEP has unemployment rising by nine-tenths, to 4.6 percent, by the end of 2023. We assess the plausibility of this projection using a specific nonlinear model that embeds an empirically successful nonlinear Phillips curve specification into a structural model, identifying it via an underutilized data-dependent method. We model core PCE inflation using three components that align with those noted by Chair Powell in his December 14, 2022, press conference: housing, core goods, and core-services-less-housing. Our model projects that conditional on the SEP unemployment rate path and a rapid deceleration of core goods prices, core PCE inflation moderates to only 2.75 percent by the end of 2025: inflation will be higher for longer. A deep recession would be necessary to achieve the SEP’s projected inflation path. A simple reduced-form welfare analysis, which abstracts from any danger of inflation expectations becoming unanchored, suggests that such a recession would not be optimal.
-
Working Paper
01.09.2023 |
WP 23-03
What drove inflation so high in 2022? Can it drop rapidly without a recession? The Phillips curve is central to the answers; its proper (nonlinear) specification reveals that the relationship is strong and frequency dependent, and inflation is very persistent. We embed this empirically successful Phillips curve – incorporating a supply-shocks variable – into a structural model. Identification is achieved using an underutilized data-dependent method. Despite imposing anchored inflation expectations and a rapid relaxation of supply-chain problems, we find that absent a recession, inflation will be more than 3 percent by the end of 2025. A simple welfare analysis supports a mild recession as preferred to an extended period of elevated inflation, under a typical loss function.
-
Working Paper
12.19.2022 |
WP 22-38
Prominent rent growth indices often give strikingly different measurements of rent inflation. We create new indices from Bureau of Labor Statistics (BLS) rent microdata using a repeat-rent index methodology and show that this discrepancy is almost entirely explained by differences in rent growth for new tenants relative to the average rent growth for all tenants. Rent inflation for new tenants leads the official BLS rent inflation by four quarters. As rent is the largest component of the consumer price index, this has implications for our understanding of aggregate inflation dynamics and guiding monetary policy.
-
Working Paper
08.03.2022 |
WP 22-23
Both theory and extant empirical evidence suggest that the cross-sectional asymmetry across disaggregated price indexes might be useful in the forecasting of aggregate inflation. Trimmed-mean inflation estimators have been shown to be useful devices for forecasting headline PCE inflation. But does this stem from their ability to signal the underlying trend, or does it mainly come from their implicit signaling of asymmetry (when included alongside headline PCE)? We address this question by augmenting a “hard to beat” benchmark inflation forecasting model of headline PCE price inflation with robust measures of trimmed-mean estimators of inflation (median PCE and trimmed-mean PCE) and robust measures of the cross-sectional asymmetry (Bowley skewness; Kelly skewness) computed using the 180+ components of the PCE price index. We also construct new trimmed-mean measures of goods and services PCE inflation and their accompanying robust skewness. Our results indicate significant gains in the point and density accuracy of PCE inflation forecasts over medium- and longer-term horizons, up through and including the COVID-19 pandemic. We find that improvements in accuracy stem mainly from the trend information implicit in trimmed-mean estimators, but that skewness is also useful. Median PCE slightly outperforms trimmed-mean PCE; both outperform core PCE. For point forecasts, Kelly skewness is preferred; but for estimating stochastic volatility, Bowley skewness is preferred. An examination of goods and services PCE inflation provides similar inference.
-
Working Paper
05.18.2021 |
WP 21-10
Core PCE inflation was designed to extract the trend in inflation from the overall PCE price index by removing transitory price changes. This paper compares core PCE inflation to trimmed mean and median PCE inflation and concludes that, for the purpose of monetary policy deliberations and communication, the latter two measures do a better job.
-
Working Paper
04.06.2021 |
WP 20-22R
Statistical agencies track rental expenditures for use in the national accounts and in consumer price indexes (CPIs). As such, statistical agencies should include late payment fees and nonpayment in rent. In the US context, late payment fees are excluded from the CPI. Ostensibly, nonpayment of rent is included in the US CPI; but its treatment is deficient, and we demonstrate that small variations in nonpayment could lead to large swings in shelter inflation, and might have played a role in the 2009 measured shelter inflation collapse. They didn’t: while the national nonpayment incidence is 2-3 percent, in the 1 million plus rent observations in BLS rent microdata from 2000-2016, no nonpayment is recorded. A back-of-the-envelope calculation suggests that, assuming nonpayment undermeasurement continued after 2016, CPI shelter inflation may have been overestimated by about 1 percentage point per month (annualized) in 2020. Late fees and nonpayment are difficult to measure in real time. We offer implementation suggestions that are consistent with CPI procedures.
-
Working Paper
02.09.2021 |
WP 21-03
Housing rents are a large share of household budgets and make a large contribution to overall inflation. Rent inflation rates for different types of housing units sometimes diverge, even in the same neighborhoods. We estimate during 2013 to 2016 apartment rents outpaced rents for detached housing in the United States by 0.76 percentage points annually after controlling for location effects. These rent dynamics imply a segmented housing market. They also suggest rent indexes need to be based on data structurally representative of their measurement objective. In particular, indexes based on professionally managed apartment complexes mismeasure the rents for housing generally. Even indexes based on careful geographical sampling, such as the Consumer Price Index’s Owners’ Equivalent Rent component, may be biased by using an unrepresentative mix of apartments and houses.
-
Working Paper
07.08.2020 |
WP 20-22
Working paper 20-22 was reviewed by staff at the BLS prior to its posting. After its posting, subsequent discussions with staff at the BLS revealed that the BLS treatment of nonpayment is different from the treatment assumed in the working paper. As a result, the authors have removed the old version of the working paper in order to incorporate this new information.
-
Working Paper
04.08.2020 |
WP 19-09R
We establish that the Phillips curve is persistence-dependent: inflation responds differently to persistent versus moderately persistent (or versus transient) fluctuations in the unemployment gap. Previous work fails to model this dependence, so it finds numerous “inflation puzzles”—such as missing inflation/disinflation—noted in the literature. Our model specification eliminates these puzzles; for example, the Phillips curve has not weakened, and inflation is not “stubbornly low” at present. The model’s coefficients are stable, and it provides accurate conditional recursive forecasts through the Great Recession. The persistence-dependent relationship we uncover is interpretable as being business-cycle-phase-dependent and is thus consistent with existing theory.
-
Working Paper
03.05.2020 |
WP 20-08
Most consistent estimators are what Müller (2007) terms “highly fragile”: prone to total breakdown in the presence of a handful of unusual data points. This compromises inference. Robust estimation is a (seldom-used) solution, but commonly used methods have drawbacks. In this paper, building on methods that are relatively unknown in economics, we provide a new tool for robust estimates of mean and covariance, useful both for robust estimation and for detection of unusual data points. It is relatively fast and useful for large data sets. Our performance testing indicates that our baseline method performs on par with, or better than, two of the currently best available methods, and that it works well on benchmark data sets. We also demonstrate that the issues we discuss are not merely hypothetical, by re-examining a prominent economic study and demonstrating its central results are driven by a set of unusual points.
-
Working Paper
08.20.2019 |
WP 19-15
Despite the stability of the median 10-year inflation expectations in the Survey of Professional Forecasters (SPF) near 2 percent, we show that not a single SPF respondent’s expectations have been anchored at the target since the Federal Open Market Committee’s (FOMC) enactment of an inflation target in January 2012, or even since 2015. However, we find significant evidence for “delayed anchoring,” or a move toward being anchored, particularly after the federal funds rate lifted off in December 2015.
-
Working Paper
07.18.2019 |
WP 18-14R
The origins of the Great Inflation, a central 20th century U.S. macroeconomic event, remain contested. Prominent explanations are poor forecasts or deficient activity gap estimates. An alternative view: the FOMC was unwilling to fight inflation, perhaps due to political pressures. Our findings, based on a novel approach, support the latter view. New econometric tools allow us to credibly identify the particular activity gap, if any, in use. Persistence-dependent unemployment (gap) responses in the 1970s were essentially the same pre- and post-Volcker. Conversely, FOMC behavior vis-à-vis inflation—also persistence-dependent—changed markedly starting with Volcker, consistent with (though not proving) the political pressures view.
-
Working Paper
05.03.2019 |
WP 19-09
We use recently developed econometric tools to demonstrate that the Phillips curve unemployment rate–inflation rate relationship varies in an economically meaningful way across three phases of the business cycle. The first (“bust phase”) relationship is the one highlighted by Stock and Watson (2010): A sharp reduction in inflation occurs as the unemployment rate is rising rapidly. The second (“recovery phase”) relationship occurs as the unemployment rate subsequently begins to fall; during this phase, inflation is unrelated to any conventional unemployment gap. The final (“overheating phase”) relationship begins once the unemployment rate drops below its natural rate. We validate our findings in a forecasting exercise and find statistically significant episodic forecast improvement. Our analysis allows us to provide a unified explanation of many prominent findings in the literature.
-
Working Paper
10.12.2018 |
WP 18-14
The historical analysis of FOMC behavior using estimated simple policy rules requires the specification of either an estimated natural rate of unemployment or an output gap. But in the 1970s, neither output gap nor natural rate estimates appear to guide FOMC deliberations. This paper uses the data to identify the particular implicit unemployment rate gap (if any) that is consistent with FOMC behavior. While its ability appears to have improved over time, our results indicate that, both before the Volcker period and through the Bernanke period, the FOMC distinguished persistent movements in the unemployment rate from other movements; implicitly such movements were treated as an intermediate target, one that departs substantially from conventional estimates of the natural rate. We further investigate historical FOMC responses to inflation fluctuations. In this regard, FOMC behavior changed in the Volcker-Greenspan-Bernanke period: its response to the inflation rate became much stronger, and it focused more intensely on very persistent movements in this variable. Our results shed light on the “Great Inflation” experience of the 1970s, and are consistent with the view that political pressures effectively limited the FOMC response to the buildup of inflation. They also suggest new directions for DSGE modeling.
-
Working Paper
05.05.2017 |
WP 17-05
Tenants’ rents are remarkably sticky even though regular recontracting would suggest substantial rent flexibility. In addition, rent stickiness varies significantly across structures such as detached units and large apartment. We offer the first theoretical explanation of rent stickiness that is consistent with these two facts. In this theory, search and bargaining with incomplete information generates stickiness in the absence of menu costs or other commonly used modeling assumptions.
-
Working Paper
12.15.2016 |
WP 16-30
This paper provides estimates of the net depreciation rate for rental housing using a unique confidential data set from the Bureau of Labor Statistics that covers over 30,000 rental units from 1998 to 2009. Our data and econometric approach allow us to add to the literature in three main ways.
-
Working Paper
12.15.2015 |
WP 15-30
We apply newly developed econometric tools (Ashley and Verbrugge, 2009a) for drawing inferences about persistence dependence in economic relationships to study the velocity of money.
-
Working Paper
11.04.2015 |
WP 15-27
We make five contributions. We demonstrate that extant trimmed-mean and median CPI construction procedures depart from Bureau of Labor Statistics index construction procedures, and that the departures don’t make much of a difference.
-
Working Paper
07.13.2015 |
WP 15-11
We study which variables correlate with rent growth in 18,000 rental units between 2001 and 2004 and 2004 and 2007. We document significant rent stickiness. Initial relative rent level is the best predictor.
-
Working Paper
11.18.2014 |
WP 14-30
We estimate a monetary policy rule for the US allowing for possible frequency dependence (the central bank can respond differently to persistent and transitory innovations) in the unemployment and inflation rates.
