About

Pascal Michaillat is a professor of economics at the University of California, Santa Cruz. His primary research focus is on economic slack, with a particular emphasis on unemployment. He investigates why unemployment exists, the reasons behind its fluctuations over the business cycle, and how macroeconomic policies should respond to these fluctuations.

Research topics

• Mathematics
• Econometrics
• Macroeconomics
• Economics
• Monetary economics

Selected publications

• Has the Recession Started?

  Oxford Bulletin of Economics and Statistics · 2025-05-09

  articleOpen access1st authorCorresponding

  ABSTRACT This paper develops a new rule to detect US recessions by combining data on job vacancies and unemployment. We first construct a new recession indicator: the minimum of the Sahm‐rule indicator (the increase in the 3‐month average of the unemployment rate above its 12‐month low) and a vacancy analogue. The minimum indicator captures simultaneous rises in unemployment and declines in vacancies. We then set the recession threshold to 0.29 percentage points (pp), so a recession is detected whenever the minimum indicator crosses 0.29pp. This new rule detects recessions faster than the Sahm rule: with an average delay of 1.2 months instead of 2.7 months, and a maximum delay of 3 months instead of 7 months. It is also more robust: it identifies all 15 recessions since 1929 without false positives, whereas the Sahm rule breaks down before 1960. By adding a second threshold, we can also compute recession probabilities: values between 0.29pp and 0.81pp signal a probable recession; values above 0.81pp signal a certain recession. In December 2024, the minimum indicator is at 0.43pp, implying a recession probability of 27%. This recession risk was first detected in March 2024.

  PublisherOA PDFDOI

• 2019 BITSS Annual Meeting

  OSF Preprints (OSF Preprints) · 2025-02-07

  other
  Publisher

• Early and Accurate Recession Detection Using Classifiers on the Anticipation-Precision Frontier

  SSRN Electronic Journal · 2025-01-01

  articleOpen access1st authorCorresponding
  PublisherDOI

• Early and Accurate Recession Detection Using Classifiers on the Anticipation-Precision Frontier

  National Bureau of Economic Research · 2025-07-01 · 2 citations

  reportOpen access1st authorCorresponding
  PublisherDOI

• Recession Detection Using Classifiers on the Anticipation-Precision Frontier

  ArXiv.org · 2025-06-11

  preprintOpen access1st authorCorresponding

  This paper develops an algorithm for detecting US recessions in real time. The algorithm constructs hundreds of millions of recession classifiers by combining unemployment and vacancy data. Classifiers are then selected to avoid both false negatives (missed recessions) and false positives (nonexistent recessions). The selected classifiers are perfect in a statistical sense: they identify all 15 historical recessions in the 1929--2021 training period without any false positives. By further selecting classifiers that lie on the high-precision segment of the anticipation-precision frontier, the algorithm delivers early detection without sacrificing accuracy. On average between 1929 and 2021, the selected classifier ensemble signals recessions 2.1 months after their true onset, with a standard deviation of detection errors of 1.8 months. The classifier ensemble is much faster than the NBER Business Cycle Dating Committee: between 1979 and 2021, the committee takes on average 6.3 months to determine recession starts, while the classifier ensemble only takes 1.2 months. Applied to September 2025 data, the classifier ensemble gives a 64% probability that the US economy has entered a recession. A placebo test and backtests confirm the algorithm's reliability.

  PublisherOA PDFDOI

• Critical Values Robust to P-hacking

  The Review of Economics and Statistics · 2024-05-06 · 5 citations

  articleOpen accessSenior author

  Abstract P-hacking is prevalent in reality but absent from classical hypothesis-testing theory. We therefore build a model of hypothesis testing that accounts for p-hacking. From the model, we derive critical values such that, if they are used to determine significance, and if p-hacking adjusts to the new significance standards, then spurious significant results do not occur more often than intended. Because of p-hacking, such robust critical values are larger than classical critical values. In the model calibrated to medical science, the robust critical value is the classical critical value for the same test statistic but with one-fifth of the significance level.

  PublisherOA PDFDOI

• Modeling Migration-Induced Unemployment

  National Bureau of Economic Research · 2024-10-01 · 1 citations

  reportOpen access1st authorCorresponding

  Immigration is often blamed for increasing unemployment among local workers. This sentiment is reflected in the rise of anti-immigration parties and policies in Western democracies. And in fact, numerous studies estimate that in the short run, the arrival of new workers in a labor market raises the unemployment rate of local workers. Yet, standard migration models, such as the Walrasian model and the Diamond-Mortensen-Pissarides model, inherently assume that immigrants are absorbed into the labor market without affecting local unemployment. This paper presents a more general model of migration that allows for the possibility that not only the wages but also the unemployment rate of local workers may be affected by the arrival of newcomers. This extension is essential to capture the full range of potential impacts of labor migration on labor markets. The model blends a matching framework with job rationing. In it, the arrival of new workers raises the unemployment rate among local workers, particularly in a depressed labor market where job opportunities are limited. On the positive side, in-migration helps firms fill vacancies more easily, boosting their profits. The overall impact of in-migration on local welfare varies with labor market conditions: in-migration reduces welfare when the labor market is inefficiently slack, but it enhances welfare when the labor market is inefficiently tight.

  PublisherDOI

• Beveridgean Phillips Curve

  arXiv (Cornell University) · 2024-01-23 · 2 citations

  preprintOpen access1st authorCorresponding

  This paper proposes a new, Beveridgean model of the Phillips curve. While the New Keynesian Phillips Curve is based on monopolistic pricing under price-adjustment costs, the Beveridgean Phillips curve is based on directed-search pricing under price-adjustment costs. Under directed-search pricing, prices respond to slack instead of marginal costs. The Beveridgean Phillips curve links the inflation gap to the unemployment gap, with the following properties. First, it produces the divine coincidence: it guarantees that the rate of inflation is on target whenever the rate of unemployment is efficient. Second, whenever the Beveridge curve shifts, the Phillips curve shifts if it is formulated with inflation and unemployment, but it remains unaffected if it is formulated with inflation and labor-market tightness. Third, the Phillips curve displays a kink at the point of divine coincidence if we assume that wage decreases -- which reduce workers' morale -- are more costly to producers than price increases -- which upset customers. These three properties describe recent US data well.

  PublisherOA PDFDOI

• Has the Recession Started?

  arXiv (Cornell University) · 2024-08-11 · 1 citations

  preprintOpen access1st authorCorresponding

  To answer this question, we develop a new Sahm-type recession indicator that combines vacancy and unemployment data. The indicator is the minimum of the Sahm indicator -- the difference between the 3-month trailing average of the unemployment rate and its minimum over the past 12 months -- and a similar indicator constructed with the vacancy rate -- the difference between the 3-month trailing average of the vacancy rate and its maximum over the past 12 months. We then propose a two-sided recession rule: When our indicator reaches 0.3pp, a recession may have started; when the indicator reaches 0.8pp, a recession has started for sure. This new rule is triggered earlier than the Sahm rule: on average it detects recessions 0.8 month after they have started, while the Sahm rule detects them 2.1 months after their start. The new rule also has a better historical track record: it perfectly identifies all recessions since 1929, while the Sahm rule breaks down before 1960. With August 2024 data, our indicator is at 0.54pp, so the probability that the US economy is now in recession is 48%. In fact, the recession may have started as early as March 2024.

  PublisherOA PDFDOI

• The Full-Employment Rate of Unemployment in the United States

  Brookings Papers on Economic Activity · 2024-09-01

  article1st authorCorresponding

  ABSTRACT: This paper computes the unemployment rate u * that is consistent with full employment in the United States. First, the paper argues that social efficiency is the most appropriate economic interpretation of the legal concept of full employment. Here efficiency means minimizing the nonproductive use of labor—both unemployment and recruiting. As it takes one worker to service one job vacancy, the nonproductive use of labor is measured by the number of job seekers and job vacancies, u + v . Through the Beveridge curve, the numbers of job seekers and vacancies are inversely related, uv = constant. With such symmetry the labor market is efficient when there are as many job seekers as vacancies ( u = v ), inefficiently tight when there are more vacancies than job seekers ( v > u ), and inefficiently slack when there are more job seekers than vacancies ( u > v ). Accordingly, the full-employment rate of unemployment (FERU) is the geometric average of the unemployment and vacancy rates: [inline-graphic 02]. From 1930 to 2024, the FERU averages 4.1 percent and is stable, remaining between 2.5 percent and 6.7 percent. Unemployment has generally been above the FERU ( u > u *), especially during recessions. Unemployment has only been below the FERU ( u < u *) during major wars, as well as shortly before and in the aftermath of the pandemic.

  PublisherDOI

Frequent coauthors

• Emmanuel Saez

  59 shared

• Camille Landais

  London School of Economics and Political Science

  11 shared

• Kristóf Madarász

  Laser Scan Engineering (United Kingdom)

  10 shared

• Erik Eyster

  University of California, Santa Barbara

  10 shared

• Adam McCloskey

  8 shared

• Andreas Stolcke

  6 shared

• George A. Akerlof

  Chatham House

  5 shared

• Jitendra Malik

  5 shared