We use cookies on our website to give you the best online experience. Please know that if you continue to browse on our site, you agree to this use. You can always block or disable cookies using your browser settings. To find out more, please review our privacy policy.


Policy Hub: Macroblog provides concise commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues for a broad audience.

Authors for Policy Hub: Macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

Comment Standards:
Comments are moderated and will not appear until the moderator has approved them.

Please submit appropriate comments. Inappropriate comments include content that is abusive, harassing, or threatening; obscene, vulgar, or profane; an attack of a personal nature; or overtly political.

In addition, no off-topic remarks or spam is permitted.

September 23, 2022

How Has the Market Responded to Restoring Price Stability?

Note: The author thanks Mark Jensen and Larry Wall for their help with this post.

The Federal Open Market Committee (FOMC) implements monetary policy chiefly through changes in its federal funds rate target, and market participants form expectations about the evolution of future monetary policy decisions based on data they think are relevant to policymakers. Between the June and July FOMC meetings, incoming data started to suggest some parts of the economy might already be feeling the effects of tighter monetary policy. On the other hand, data since July tell a different story, one that suggests the FOMC still has a way to go in its efforts to fight inflation and restore price stability. So how have market participants interpreted these disparate pieces of data, and what could they mean for future monetary policy decisions?

In this post, I use the Atlanta Fed's Market Probability Tracker to understand how data since the June and July FOMC meetings have affected the market's expectations about the path of future monetary policy, similar to the analysis I did with Atlanta Fed economist Mark Jensen in a Macroblog post late last year. As another Atlanta Fed colleague, Mark Fisher, and I discussed in a Notes from the Vault article, the Market Probability Tracker generates estimates of the expected federal funds rate path based on eurodollar futures and options on eurodollar futures. Eurodollar futures and options deliver a three-month LIBOR (the London interbank offered rate) interest rate average, which is closely linked to the federal funds rate. Additionally, eurodollar futures and options are among the most liquid of financial instruments, with contracts that are expiring several years in the future regularly traded. As a result, the Market Probability Tracker generates our best estimates of expected rate paths by incorporating all the available data that the market believes will affect future policy decisions.

Figure 1 below uses the expected rate paths produced by the Market Probability Tracker to illustrate how market expectations between the June and July FOMC meetings responded to new information about the economy. The solid black line in figure 1 below shows the federal funds rate path expected by market participants after the FOMC's meeting press conference on June 15 video fileOff-site link. The black dotted and dashed lines represent, respectively, expectations after Fed chair Jerome Powell's Senate Banking Committee testimony on June 22Off-site link and after the release of the US Department of Labor's weekly report of unemployment insurance initial claims Adobe PDF file formatOff-site link on June 30. Although economic updates occurred throughout the intermeeting period (the New York Fed maintains a list of important releases going back to 2018 in its Economic Indicators Calendar), the changes in expectations on these two dates best summarize the overall change in the market's expectations. Lastly, the solid orange line represents expectations following the FOMC's press conferenceOff-site link on July 27. I also include (shaded in gray) the target range of 225 to 250 basis points announced at that meeting.

Starting with the black dashed line, after the June FOMC meeting, market participants expected the federal funds rate target range to reach 375 to 400 basis points by the first quarter of 2023 and then fall 75 basis points over the course of the next two years. During his June 22 Senate Banking Committee testimony, Chair Powell said that an economic downturn triggered by rate hikes to tame inflation was "certainly a possibility," although he added that such a downturn was neither the Fed's intent nor, in his view, necessary. His statement was important, albeit qualitative, information for the market because the FOMC in the past has often responded to economic downturns by lowering interest rates, which market participants interpreted as lower overall federal funds rates in the future (represented by the dotted black line). Expectations fell even lower after the Department of Labor's June 30 report hinted at a moderating labor market, with initial claims near five-month highs (represented by the dashed black line). By the July FOMC meeting (represented by the orange line), market participants further expected rate hikes to end this year and then fall 100 basis points during the next two years, as subsequent initial unemployment claims reports remained elevated and the July 21 Philadelphia Fed's manufacturing survey showed a drop in activity.

Figure 2 shows the Market Probability Tracker's estimates of the federal funds rate path expected by market participants following the July FOMC meeting (the solid orange line). It also shows the expected rate path following the August 5 release of the July employment situation reportOff-site link from the US Bureau of Labor Statistics (represented by the dotted black line), the expected rate path following the September 8 release of the initial unemployment claims report (represented by the dashed black line), and the expected rate path following the September 13 release of the August consumer price indexOff-site link (represented by the dot-dash line). Much like figure 1, these dates best summarize how market participants reacted to the evolving data since the July FOMC meeting (despite many other data releases occurring throughout the intermeeting period). Lastly, the solid black line represents the expected rate path on September 20, the day before the press conference following the FOMC meeting.

In the eight weeks since the July FOMC meeting, the data that the Committee said it would use to evaluate future policy moves came in much stronger than expected. Rather than moderating, labor markets appeared to tighten, with the Bureau of Labor Statistics reporting a 526,000 increase in nonfarm payrolls in its July jobs report and the Department of Labor reporting a decline in initial unemployment claims throughout August that culminated, in the September 8 report, in their lowest reported levels since May. The consumer price index for August, which many had expected to fall on a month-over-month basisOff-site link, showed inflation increasing 0.1 percent from July. The less-volatile core measure of inflation that excludes gasoline and food prices also rose 0.6 percent from July, twice the expected rate. Given the data's direction leading up to the September FOMC meeting, market participants expected a more aggressive pace of rate hikes through the end of the year, from 325 basis points after the July FOMC meeting to nearly 450 basis points. They also expect rates to remain much higher for much longer, with rates at the end of 2025 near 350 basis points—which would be at least 100 basis points higher than the 225 to 250 basis points that the chair described as the "neutral" policy rate at his July press conference.

Figure 3 shows the Market Probability Tracker's estimates of the federal funds rate paths that market participants expected the day before the press conference following September's meeting (the solid black line), and after the press conference on September 21 Adobe PDF file formatOff-site link (the solid orange line). Chair Powell, in his opening remarks, commented that tight labor markets "continue to be out of balance" with demand and that "price pressures remain evident across a broad range of goods and services." The information contained in both the press conference and the material released by the Committee did not significantly change market expectations about the future path of monetary policy, which already incorporated recent data on inflation and labor market conditions.

Turning back to the question posed by this post's title, the rate path movements seen in reaction to the incoming data show that, initially, market participants expected rate hikes to end in 2022. But the data, which came in much stronger after the July FOMC meeting, led the market to expect the Fed to raise rates higher than had been expected following the June FOMC meeting. Market participants also expect the Fed to keep those rates much higher for longer in order to cool demand—as Chair Powell put it in his August 26 speechOff-site link at the Jackson Hole economic policy symposium, "until we are confident the job is done."

More importantly, the movements in the rate paths highlight the insights we can gain from the Market Probability Tracker into how information about the economy affects the market's expectations of future monetary policy decisions. Chair Powell observed during the June press conference that "monetary policy is more effective when market participants understand how policy will evolve." With the rate paths produced by the Market Probability Tracker each day, we can begin to make that assessment.

August 31, 2022

Lessons from the Past: Can the 1970s Help Inform the Future Path of Monetary Policy?

People in monetary policy circles sometimes use the phrase "long and variable lags" to describe the delayed impact of the Fed's main policy tool on demand and inflation. The popularization of that phrase can be traced to a speech by Milton Friedman during the 1971 American Economic Association meetings, and since then people usually use it to describe the impact of Fed policy on economic output and inflation. Yet, during that speech, when summing up his work on the subject, he noted that "...monetary changes take much longer to affect prices than to affect output," adding that the maximum impact on prices is not apparent for about one and a half to two years.

Since Milton Friedman, many economists have studied these "long and variable lags" (including former Fed chair Ben Bernanke). And, while the length of the lag has proven "variable" as first suggested, the main result still rings true. Changes in the stance of monetary policy have the largest impact on output first and then, much later, on inflation. A large literature bears out this assertion. Bernanke et al. (1999) and Christiano, Eichenbaum, and Evans (2005) point to a two-year lag between monetary policy actions and their main effect on inflation. Gerlach and Svensson (2001) report an approximately 18-month lag in the euro area, while Batini and Nelson (2001) estimate that changes in the money supply have their peak impact on inflation in the UK after a year.

That context is especially useful for monetary policymakers to keep in mind as they navigate the economic challenges of the pandemic. In a span of just two and a half years, the US economy has suffered its sharpest post-WWII decline in economic output, a subsequent rapid resurgence in demand, a dramatic disconnect between labor supply and labor demand, widespread supply and shipping constraints, and an inflation rate that has surged from roughly 1.5 percent to 9 percent in the past 17 months. And, despite current strong job growth and the highest inflation this country has seen in 40 years, worries over a potential recession mount (as evidenced by the number of questions Chair Powell was asked about the "r word" in his press conference Adobe PDF file formatOff-site link following the most recent meeting of the Federal Open Market Committee). These beliefs partly reflect the rapid shift in the fiscal and monetary policy stance over the past year. In response to the pandemic, Congress approved a stimulus package of $5 trillionOff-site link, while the Fed expanded its balance sheet by roughly the same amount. But now, the federal deficit has fallen more than 81 percentOff-site link in first 10 months of 2022 fiscal year compared to 2021. In turn, the Fed has embarked on policy normalization, raising interest rates well into the range of neutral and drawing down its balance sheet (actions known as quantitative tightening).

Although "this time is different," we might be able to gain insights into the appropriate path of monetary policy by revisiting the past. At the time of Milton Friedman's 1971 speech, the economy was coming out of what many economists saw as a policy-driven recessionOff-site link , which followed a period of fiscal tightening to make up for large government outlays for the Vietnam War and a sizeable slowing in money growth as the Fed attempted to quell rising inflation. Today, the main policy tool is the federal funds rate, but prior to the early 1980s, changes in the money supply were the primary instrument. (Monetary aggregates—that is, growth in the money supply—formally replaced bank credit as the primary intermediate target of monetary policy in 1970. At the time, the fed funds rate played only a secondary role and was used as guideline in day-to-day open market operations, aimed at smoothing short run volatility.) In the run-up to the 1969–70 recession, the Fed tightened policy, slowing the growth in the money supply from 8 percent on a year-over-year basis to just 2 percent (the associated increase in the fed funds rate was roughly 4.5 percentage points, to 9 percent). Yet, as quickly as the Federal Open Market Committee tightened policy in the late 1960s, it more than reversed course in response to a sizeable increase in the unemployment rate during the recession. By late 1971, the money supply was surging again, up 13 percent on a year-over-year basis.

The Fed's quick and stark policy reversal became a recurring theme in the 1970s. During the decade, the Fed quickly pivoted between battling high unemployment and high inflation, what many economists refer to as "stop-and-go" policiesOff-site link. Charts 1 and 2 clearly show these shifting stances as they occurred again in the run-up to and aftermath of the 1974–75 recession. Chart 1 plots the year-over-year growth rate in the money supply (M2) and the unemployment rate, and chart 2 plots the growth in the money supply against the year-over-year growth rate in consumer price index inflation.

These charts depict three points that remain salient today. First, the "stop-and-go" policies of the 1970s clearly highlight the "long and variable lags" that changes in monetary policy have on inflation. Money growth plateaus at high levels three times during the late 1960s through the 1970s: in 1968, 1972, and in the mid-1970s. Each of those periods is followed by a subsequent surge in inflation, prompting a sustained tightening of monetary conditions. But as soon as inflation began falling, the Fed quickly reversed course with a bold expansion in the money supply that overshadowed the one originating the previous cycle, citing spikes in unemployment along with a lagged decline in inflation as justifications for these reversals.

If we smooth through some of the cyclical dynamics, there was a sustained upward drift in both the unemployment rate and inflation. In the mid-1960s, both inflation and the unemployment rate were around 2 percent. And 1980 inflation was over 10 percent and the unemployment rate had drifted up to 6 percent.

This period's upward drift in unemployment and inflation ran counter to the era's prevailing wisdom, which held that higher inflation was simply the sacrifice needed to lower the unemployment rate, and vice versa, An insightful essayOff-site link by former Atlanta Fed economist Mike Bryan covers this period in depth. He writes, "The stable trade-off between inflation and unemployment proved unstable. The ability of policymakers to control any ‘real' variable was ephemeral. This truth included the rate of unemployment, which oscillated around its ‘natural' rate. The trade-off that policymakers hoped to exploit did not exist."

Why didn't this stable tradeoff exist? Part of the answer is that the unemployment rate fluctuates around an unobserved natural rate. (Fed chair Jerome Powell's 2018 speechOff-site link offers an accessible discussion of these unobservables.) But the other part of the answer that is particularly salient at the moment is that by the mid- to late 1970s, after enduring a sustained period of rising unemployment and inflation, people began to expect higher inflation rates.

Chart 3 plots one-year-ahead inflation expectations alongside inflation and money growth using data from the Livingston SurveyOff-site link, a twice-annual survey of a small group of professional economists that the Philadelphia Fed has conducted since the end of WWII. And here the upward drift in inflation expectations is striking. By 1980, inflation expectations had risen 10 percent. Our interpretation of these data is that the rapid reversals of policy that characterized Fed actions during the 1970s never allowed inflation to fall back to the 1 to 3 percent range that was the norm after the end of the Korean War. As a consequence, the expectation that inflation would not recede into the background eventually became embedded into the psyche of Americans. People who lived through this experience simply anticipated higher future inflation rates, with that expectation embedded into their price-setting and wage-bargaining decisions.

Now, history here is messy. A number of caveats and confounding factors contributed to the unfavorable economic outcomes of the 1970s. Fed historians such as Allan Meltzer argue that the prevailing Fed chair at the time, Arthur Burns, did not consider monetary policy as ultimately responsible for such high inflation. Instead, the chair pointed to unions' wage-bargaining power first and, in particular, "cost-push" shocks (that is, energy and food shortages) later as the responsible party. (And indeed, this "cost-push" theory of inflation, so prevalent at the time, merits further exploration since assuming that spikes in energy prices might have contributed to the unanchoring of inflation expectations makes sense.)

Yet, in the case of oil price shocks, there is a counterpoint. The breakdown of the Bretton Woods accords ultimately drove the sustained increase in oil prices, and their breakdown can be seen in the era's robust money growth at the time. The breakdown of these accords created a run on the dollar amid fears of inflation. The price of gold took off as many investors were scrambling for an inflation hedge. Interestingly, the increase in the price of oil actually followed the spike in the price of gold and other commodities.

The historical evidence suggests that by 1970, the attempt to defend the dollar at a fixed peg of $35 per ounce, established by the Bretton Woods agreements, had become increasingly untenable, and gold outflows from the United States accelerated amid sustained inflation and trade/fiscal deficits. The run on the dollar forced President Nixon to effectively "close the gold window," making the dollar inconvertible to gold in August 1971. One month later, OPEC communicatedOff-site link its intention to price oil in terms of a fixed amount of gold. Hence, the increase in the money supply, spurred by the run-up in gold prices, exacerbated the increase in the dollar price of oil and led to the high inflation that followed. OPEC was slow to readjust prices to reflect this depreciation. However, the substantial price increases of 1973–74 and 1979 largely returned oil prices to the corresponding gold parity (see chart 4), which, again, was then seen as an inflation hedge.

In this context, it's worth noting that the OPEC oil embargo following the Yom Kippur WarOff-site link lasted just a few months, but the price increase was permanent. Similarly, the drop in oil production following the Iranian Revolution was negligible, as Saudi Arabia increased production to offset most of the decline. Contrast these episodes to the Gulf War in 1990. Oil prices doubled during the conflict (July–October) but went back to previous levels once the war ended.

In sum, Arthur Burns leaned heavily into the notion that these cost-push shocks—and not Fed action—were ultimately responsible for inflation, effectively ignored the "long and variable" lags of monetary policy, misread the monetary dynamics, and reacted expediently to the real-side damage that high energy prices wreaked on the economy.

So let's fast-forward to today. The fiscal response to the onset of the pandemic was quite forceful—$5 trillion by most countsOff-site link—and at least on par with significant wartime spending. As these transfers and disbursements hit households' wallets and businesses' ledgers, money growth surged higher than 25 percent—peaking well above, though not as sustained as, the high money-growth periods during the 1970s (see chart 5). And we've seen a sharp surge in inflation that has gone well beyond pandemic-related supply constraints and shipping bottlenecks that affected certain production inputs such as computer chips. As of July 2022, roughly three-quarters of consumers' market basket rose at rates in excess of 3 percent (and two-thirds of the market basket increased at rates north of 5 percent). These levels are on par with those we saw during the Great Inflation of the 1970s.

The Committee has begun an aggressive campaign to squelch this inflation threat, hiking rates in each of the last four meetings by a cumulative 2 percentage points along with implementing plans to reduce the size of the Federal Reserve's balance sheet. It has also indicated that more policy tightening is to comeOff-site link. If history is any guide, at least in broad strokes, it will take some time before recent policy actions begin to affect inflation.

And here, it appears that the FOMC is very attuned to the lessons from the Great Inflation period. In a recent speech at Jackson HoleOff-site link, Chair Powell noted, "Our monetary policy deliberations and decisions build on what we have learned about inflation dynamics both from the high and volatile inflation of the 1970s and 1980s, and from the low and stable inflation of the past quarter-century." Perhaps more importantly, he emphasized, "Restoring price stability will likely require maintaining a restrictive policy stance for some time. The historical record cautions strongly against prematurely loosening policy."

August 2, 2022

Firms' Inflation Expectations: Not Unanchored, but Perhaps Unsettled?

Last week, the Federal Open Market Committee (FOMC) again decided to raise the federal funds rate target by 75 basis pointsOff-site link in an effort to help curb inflation. While Chair Powell, in his press conference following the July FOMC meeting, noted the lag in monetary policy's influence on inflationary pressures, he also pointed out that "if you have a sustained period of supply shocks, those can actually start to undermine or to work on de-anchoring inflation expectations" (see the 16-minute mark of the video video fileOff-site link from the press conference).

We, too, share this worry. In fact, it's been something we've been concerned about here at the Atlanta Fed since last fall. In a speech at Peterson Institute speech in October 2021, Atlanta Fed president Raphael Bostic said: "I continue to believe currently elevated inflation is episodic, driven by pandemic conditions such as disruptions in supply chains and labor markets. A major caveat, though, is that the severe and pervasive supply chain issues will probably last longer than most of us initially expected. Up to now, indicators do not suggest that long-run inflation expectations are dangerously untethered. But the episodic pressures could grind on long enough to unanchor expectations."

After monitoring the evolution of our Business Inflation Expectations (BIE) survey data for the better part of a year, we're still concerned. While the BIE has only been around since 2011, and the current period of high inflation is the only shock we've picked up in our data, the speed and sharpness of the increase in our survey-based measure of inflation expectations have left us wondering: How unsettled are longer-run inflation expectations?

For those not familiar with the BIE survey and its unique perspective on firms' inflation expectations, it's a monthly regional survey that the Atlanta Fed conducts to measure firms' inflation expectations in the Sixth District. Rather than elicit firms' aggregate inflation expectations, we ask them about their own-firm unit cost expectations—an input that is directly related to their own-firm price formation strategies (see Meyer et al. 2021). Moreover, the expectations we elicit are probabilistic, meaning we can gauge not only an individual firm's average expectations but also the level of certainty with which firms are holding those views. What does the BIE tell us about how firms are reacting to the highest bout of inflation the United States has experienced in 40 years?

By now, the current inflationary environment should be apparent to just about everyone. It's affecting all areas of the consumer market basket and has become a recurring headline in the news. It's even become a popular search term on Google. In fact, inflationary pressures have risen to heights that we haven't seen since the early 1980s, during the period known as the Great Inflation. As the chart below shows, roughly three-quarters of prices in the consumers' market basket (by expenditure weight in the consumer price index, or CPI) have risen at rates equal to or exceeding 5 percent (see chart 1). In June, more than 90 percent of prices in the market basket outpaced the 3 percent mark.

This high-inflation environment is not lost on businesses. In fact, firms' perceptions (unit-cost realizations) have been highly correlated with the evolution of overall inflation over the course of the pandemic. Chart 2 plots firms' realized unit-cost growth over the past year and compares it to the year-over-year growth rate in overall inflation as measured by the GDP price index, which is the broadest measure of inflationary pressures as it goes beyond just consumer prices to gauge price pressures in the entire economy.

In the world of survey-based measures of inflation expectations, our BIE survey has one clear advantage: not only can it track the evolution of expectations but it can also provide clear insight into firms' perceptions of the current environment, which is very useful in gauging the external validity of those expectations. To us, this insight compellingly shows that firms in our BIE panel are clearly seeing the facts when it comes to inflation.

When general prices increase, businesses' input costs also increase, and as higher costs squeeze margins, many firms will pass some or all of those costs on to their customers in the form of (you guessed it) higher prices. Survey evidence suggests a correlation between supply chain disruptions and higher year-ahead inflation expectations. We've previously noted that although supply chain disruption isn't the only factor influencing expectations, firms with the largest levels of disruption tend to hold higher expectations for inflation in the year ahead (see chart 3). So what are firms telling us about their expectations for the evolution of inflation over the year ahead and beyond?

Chart 3 of 4: Firms' Year-Ahead and Longer-Run Expectations Have Increased

Perhaps the easiest way to see how much both short-run and five-year-ahead (long-run) inflation expectations have moved over the past two years is to index them to their prepandemic growth rates. In chart 3, we depict these expectations, indexing each series to 100 in the fourth quarter of 2019. What we can see is that both short-run and longer-run expectations have increased dramatically.

While initially short-run expectations dipped at the start of the pandemic (amid widespread lockdowns and a substantial decline in economic activity), starting early in 2021, firms' short-run inflation expectations began to increase sharply. Later in 2021, firms' long-run expectations also started to show a meaningful pickup. Firms' short-run inflation expectations have roughly doubled relative to the prepandemic period. And firms' longer-run expectations are about 25 percent higher than the expectations we saw in late 2019.

We can dig a bit deeper into firms' longer-run expectations by examining the average probability weights that firms assign to the potential outcomes for longer-run unit costs at different periods, as chart 4 shows. In this case, we look at the fourth quarter of 2019 and the second quarters of 2020, 2021, and 2022.

These histograms illustrate the degree to which firms' inflation expectations have shifted over the course of the pandemic. Here, two aspects of this shift stand out to us. First, through the middle of 2021, even as inflation metrics were beginning to heat up, the distribution of firms' longer-run expectations hadn't moved much. Second, during the past year, the average probability distribution shifted starkly. The typical firm in our panel is now assigning nearly a 60 percent probability to longer-run unit cost increases of at least 3 percent per year. Moreover, their modal expectation is for longer-run unit costs to increase by 5 percent or more annually—which means that more firms anticipate the need to adjust their unit costs by more than 5 percent per year in the long run. To put this bluntly, we haven't witnessed anything like this in the decade-long existence of this survey.

A couple of caveats are worth mentioning here. First, although this is the first sizable "inflation shock" we've been able to examine in the BIE survey, we do not have a long enough time series to compare the current era to the aforementioned Great Inflation. At best, we can suggest that—given the high correlation between firms' unit-cost expectations and professional forecasters' expectations—our measures would have performed similarly in the '70s and '80s. Also, as the extensive literature on consumer expectations documents, the possibility exists for business executives to base their projections for future unit costs solely on current conditions (a phenomenon economists call adaptive expectations). Still, that last point cuts two ways. First, it's possible that, should inflation ebb meaningfully in the coming months, these longer-term expectations might follow suit. Conversely, persistently high inflation could further cement such expectations for the longer run, making it more challenging for policymakers to bring inflation back to their price-stability goals.

Said another way, the current bout of high inflation is unusual in many different ways, and how it will play out remains fraught with uncertainty. Firms' short- and long-run expectations have risen sharply, and longer-run expectations show a clear rise in the average firm's probability distribution, to the extent that nearly one-third of the weight is being assigned to anticipated cost increases greater than 5 percent. So as we continue to delve further into these expectations and monitor upcoming developments, we're left pondering the question: is this how "unanchoring" begins?

July 13, 2022

Rounded Wage Data and the Wage Growth Tracker: An Update

In an earlier Policy Hub: Macroblog post, I noted that the US Census Bureau had announced that it planned to make changes to the Current Population Survey Public Use File (CPS PUF). Those changes, part of the Enhanced Disclosure Protection program, included the rounding of the reported wage data in a way that would have a dramatic impact on the usefulness of the Atlanta Fed's Wage Growth Tracker.

The Census Bureau subsequently revised its plans and has proposed a different rounding method described here, to be introduced in February 2023 Adobe PDF file formatOff-site link. This Macroblog post looks at the new method's potential impact on the Wage Growth Tracker. It also considers another of the Census Bureau's other proposed changes to the CPS PUF.

table 01 of 01: Current proposed rounding rules

So, for example, $19.99 an hour would become $20, whereas $19.95 an hour would be unchanged. Also, $999 a week would be rounded to $1,000, while $995 would be unchanged.

How much impact would this revised scheme have had on the Wage Growth Tracker if it had been used in the past? The following chart plots three versions of the Wage Growth Tracker time series. The blue line is the published Wage Growth Tracker using unrounded data. The gray line is the Tracker based on the original proposal and described in the earlier Macroblog post. The orange line is the Tracker based on the revised rounding rules. The following table summarizes the current proposed rounding rules:

chart 01 of 01: three-month moving average of the median 12-month growth in the hourly wage of matched wage and salary earners

The chart makes clear that the impact on the Wage Growth Tracker under the current proposed method for rounding is much smaller than the original proposal. While the revised method holds some impact, the basic time series properties of the historical Wage Growth Tracker remain largely intact. The largest difference between the Wage Growth Tracker based on the current proposal and the Tracker computed using unrounded wage data is 0.13 percentage points, the average difference is −0.002 percentage points, and the mean absolute difference is 0.03 percentage points.

No approach is perfect, though, and one quibble I have with the current proposal is that the rounding schemes for reported hourly and weekly wages are not very consistent. For example, for someone who usually works 40 hours a week (the most commonly reported workweek), rounding an hourly wage less than $20 to the nearest $0.05 should be the same as rounding a weekly wage less than $800 to the nearest $2. But the current proposal rounds a weekly wage less than $800 to the nearest $5 instead. For someone reporting a wage of between $20 and $39.99 an hour, the proposed rounding to the nearest $0.25 equates to rounding a 40-hour weekly wage between $800 and $1,599 to the nearest $10. However, the current proposal rounds a weekly wage between $800 and $1,000 to the nearest $5, and a weekly wage above $1000 to the nearest $25. Finally, for someone reporting an hourly wage of $40 or more, the proposed rounding to the nearest $0.50 equates to a 40-hour weekly wage of $1,600 or more rounded to the nearest $20. But the proposal rounds a weekly wage of $1,600 or more to the nearest $25.

The preceding analysis suggests that a more consistent method would be to round a weekly wage less than $800 to the nearest $2, a weekly wage between $800 and $1,599 to the nearest $10, and a weekly wage above $1,600 to the nearest $20.

This alternative rounding method reduces the impact on the Wage Growth Tracker series relative to the current proposal by about one-third. Specifically, the mean absolute difference between the unrounded Tracker series and the series based on the currently proposed rounding scheme is 0.03 percentage points, versus 0.02 percentage points using my alternative. The largest difference is 0.09 percentage points, and the average difference is 0.002 percentage points.

For the CPS PUF, the current proposal has another aspect relevant to the Wage Growth Tracker: the future computation of topcoded earnings data. Currently, a threshold hourly wage that varies with hours worked is used to determine if an hourly wage is topcoded. For weekly earnings the threshold is $2884.61 ($150,000 a year). However, these threshold values have not changed since 1998, and because of generally rising nominal wages over time this has led to the topcoding of more wage observations each year (see hereOff-site link for more discussion of this issue). The Wage Growth Tracker's calculations exclude topcoded wage values because their inclusion would be computed as zero wage change—artificially pulling median wage growth lower.

The current proposal would instead compute a dynamic topcode value that varies in a way that would result in the top-coding of only the highest 3 percent of earnings each month. Although that change means more observations to use to compute the Tracker, those observations will come from a part of the wage distribution that might exhibit quite distinct wage growth properties. For example, wage growth tends to be lower for people at the end of their careers than at the start, and if the highest wages are mostly from people with relatively low wage growth, median wage growth could be pulled lower. Unfortunately, without access to the historical wage data that are not topcoded, constructing a counterfactual to explore the impact of this proposed change is simply not possible. Perhaps someone at the Census Bureau will explore the impact this change has on the properties of the wage growth distribution.

The Census Bureau is seeking comments on the Enhanced Disclosure Protection proposals through July 15, 2022. If you have any suggestions on any aspect of the proposal, send an email to ADDP.CPS.PUF.List@census.gov. I will be sending them a copy of this post for their consideration.