How Data Revisions Shape the Federal Reserve's Policy Decisions

The Federal Reserve has long championed a data-dependent approach, anchoring its crucial decisions on the solid ground of empirical evidence.

by Faruk Imamovic
How Data Revisions Shape the Federal Reserve's Policy Decisions
© Getty Images/Anna Moneymaker

The Federal Reserve has long championed a data-dependent approach, anchoring its crucial decisions on the solid ground of empirical evidence. However, the reliability of this data, once considered a cornerstone of informed decision-making, is now under scrutiny.

As Federal Reserve officials grapple with the pivotal decision of adjusting interest rates, a fundamental question arises: What happens when the data they rely on is not as steadfast as it once seemed? This emerging conundrum is not just a theoretical concern but a practical challenge that is reshaping the landscape of monetary policy.

Fed Governor Christopher Waller's candid admission late last year sheds light on this dilemma. "We have to make decisions in real time," he stated, acknowledging the limitations imposed by the timing and reliability of economic data.

The crux of the issue lies in the fact that data is not static; it undergoes revisions, often significantly altering the economic narrative months after initial reports are released.

The Unpredictability of Economic Data

In the realm of economic policymaking, the accuracy and timeliness of data are of paramount importance.

Fed Governor Christopher Waller's reflections underscore a key challenge: the necessity of making decisions based on the data available at the moment, despite its potential for later revision. This issue becomes particularly thorny when these revisions are not minor adjustments but game-changers, significantly altering the economic landscape perceived by policymakers.

Take, for example, the interpretation of the job market in 2021. Initial monthly headline employment numbers painted a picture of a job market that was "okay, but not really great," according to Waller. This perception, coupled with the highest inflation rates in four decades, led to a cautious approach towards raising interest rates, fearing potential job losses.

However, this initial understanding was turned on its head by subsequent data revisions. The case of the August 2021 jobs report is illustrative. Initially, the Bureau of Labor Statistics (BLS) estimated that 235,000 new jobs were added.

However, this figure underwent significant revisions. The first two revisions alone, incorporated in the following two months' reports, nearly doubled this number to approximately 483,000. Waller's reflection on this situation is telling.

“When you look back, you’re like, 'Oh my God, the labor market was a lot stronger than it was indicated by the release of the data at the time,’” he remarked. This hindsight perspective highlights a critical issue: in real-time, data may not accurately reflect the economic reality, leading to potentially delayed or misinformed policy decisions.

Christopher Waller© Getty Images/Sarah Silbiger

The Ripple Effect of Data Revisions

The phenomenon of data revision plays a significant role in shaping economic policy, as evidenced in both 2021 and the following years.

In 2021, the labor market appeared weaker than initially reported, with revised employment gains for all but one month indicating a less robust job market. This trend of significant data revisions continued into the following year, presenting a challenge for policymakers.

Fed Governor Michelle Bowman highlighted the difficulties arising from frequent and substantial data revisions. Speaking at an event hosted by the Ohio Bankers League in November, Bowman pointed to recent revisions in monthly job gains and average hourly earnings, emphasizing the complexities these adjustments introduce in forecasting economic trajectories.

The Federal Reserve's own meeting summaries, known as the Fed minutes, echoed this sentiment. In one meeting, several participants noted the challenges in assessing the state of the economy due to volatile data subject to large revisions.

This uncertainty complicates the Fed's efforts to steer the economy, especially in times of rapid change. The revisions in Gross Domestic Product (GDP) data further illustrate this point. In early 2023, the Commerce Department's initial estimate indicated that the US economy grew at an annualized rate of 1.1% in the first quarter, significantly below the 2% rate forecasted by economists.

This data, albeit preliminary, would have suggested to Fed officials that their efforts to rein in inflation by slowing economic growth were bearing fruit. However, subsequent revisions painted a different picture. A slight increase to 1.3% was reported a month later, followed by a final revision aligning with the original 2% growth forecast.

The Challenge of Accurate Data Collection

Accurately capturing the pulse of a nation's economy through data is a formidable task, fraught with inherent challenges. This is particularly true in the case of economic indicators, where data collection relies heavily on surveys and estimations.

The nature of these methods brings with it a degree of imperfection, echoing the limitations often seen in election polls where predictions don't always align with outcomes. For economic data, the crux of the issue lies in the methodology.

Government agencies like the Bureau of Labor Statistics (BLS) and the Census Bureau rely on surveys to extrapolate estimates representative of the entire economy. The accuracy of these estimates hinges on the size and diversity of the survey sample.

The larger and more diverse the sample, the closer the estimate is likely to mirror reality. However, post-pandemic, there has been a notable decline in survey response rates. This drop poses a risk of bias in the data collected, potentially skewing economic reports.

Erica Groshen, a former commissioner of the BLS, points out that while the agency rigorously tests for known biases, the decreased response rates might introduce new, undetected biases. David Wilcox, a former Fed staff member and current economist at the Peterson Institute for International Economics and Bloomberg Economics, highlights a specific concern.

He suggests that firms participating in BLS surveys might differ significantly from those that do not, potentially leading to a misrepresentation of the economic conditions. This issue of survey participation and response rates is not solely a statistical concern.

Laura Kelter, from the BLS, acknowledges the challenge, citing various factors contributing to declining response rates. These include the voluntary nature of participation, survey fatigue, and increasing concerns about confidentiality and data security.

These challenges are compounded by societal changes, such as lessened civic responsibility and the prevalence of technologies like caller ID and spam filters.

Navigating the Economic Landscape with Evolving Data

The challenge for the Federal Reserve and other economic bodies will be to continue adapting their strategies to the evolving landscape of data.

In doing so, they must balance the immediacy of real-time decision-making with the retrospective clarity that data revisions provide. The goal remains clear: to navigate the complexities of the economy with an informed, pragmatic approach, ensuring that policy decisions are as grounded and effective as possible in an ever-changing world.

Federal Reserve