U.S. Economic Snapshot

By Paul Gomme and Peter Rupert

The BLS announced the CPI and PPI for June. The CPI rose 0.18% from May to June on a seasonally adjusted basis, 2.18% when annualized, and was up 3.1% year over year. Rather than strip out a lot of goods that comprise the CPI, we look at the CPI on a three month average basis. As shown in the graph, the year over year continues to decline while both the annualized monthly and the 3 month ticked up slightly. The CPI-X (ex food and energy) differs, however. The annualized monthly rate grew 1.91%, below the Fed’s 2% target.

Of course, the question the Fed faces comes down to: What has been happening to the trend in inflation? Given the 2% target, the Fed hopes to see progress toward that goal. But, as we have commented on in several previous posts, what measure of inflation is the best indicator? Headline CPI (PCE)? Core CPI (PCE)? The Fed’s new Supercore CPI (PCE) (the price of services excluding energy and housing services)? Paul Krugman (NYT, July 11) has his own measure of supercore: “My preferred measure these days is “supercore,” which excludes food, energy, used cars and shelter (because official measures of housing costs still reflect a rent surge that ended a year ago.)” And goes on to say he is not a fan of the Fed’s measure, “The Fed has a different measure of supercore — non-housing services — but when you look at the details of that indicator, it’s a dog’s breakfast of poorly measured components that I find hard to take seriously.” Here is a graph of the Fed’s supercore measure:

So what are we to make of all of this? The main problem is that all of these measures are purely arbitrary; there is no underlying theory as to how, or what to measure. The BLS states, “Inflation can be defined as the overall general upward price movement of goods and services in an economy. The U.S. Department of Labor’s BLS has various indexes that measure different aspects of inflation.” These new so-called supercore measures omit various categories of goods that are purchased in the market. The overall CPI is comprised of 243 commodities and services measured in 32 different geographic areas. To show just how big these omissions are, the following table recreates a subset of the BLS table where the relative importance is the share of total expenditure used to weight the respective change and the May 2023-June 2023 column gives the monthly change:

The table above shows that the core CPI includes about 79.7% of all expenditures. Removing shelter from the core means the price measure includes only 45% of the things we buy. If only services prices are included that would vastly reduce the set of prices used to calculate inflation.

The BLS announced that the Producer Price Index (PPI) for final demand increased 0.1% from May to June or 1.65% annualized. Year over year it increased 0.24%. Our 3 month measure fell -0.69%, the third consecutive month with a decline.

Trend Inflation

What we (the writers of this blog) mean by “trend inflation” is a series that tracks the movements of actual monthly inflation but somewhat smoothed (taking out the more extreme fluctuations). As we’ve discussed before, this is a signal-extraction problem. Every month, we get a new CPI number from which we want to make an inference concerning the current trend inflation rate. Trend inflation is the “signal”. However, the news (the new CPI number) is a noisy indicator of this signal, both because the CPI is reported in levels and because of transitory factors affecting monthly inflation. Further, the nature of the data is such that we cannot easily extract the signal (trend inflation) from the data. With all of this in mind, the reason for looking at core or supercore CPI inflation is that it has a higher signal-to-noise ratio than CPI inflation. Both seem promising in that they omit the more volatile price components of CPI: in the case of core CPI, food and energy; in the case of super core CPI, lots of other stuff. However, the charts above show that over the past couple of years, core CPI inflation has been running well above actual CPI inflation while the Fed’s supercore inflation has been well below actual.

An alternative to slicing and dicing CPI is to compute inflation over a horizon longer than a month. The idea in this case is that the monthly noise is an independent draw (in the jargon of economics, it’s independently and identically distributed) while trend inflation has memory and changes relatively slowly over time. Consider headline inflation. Typically, it’s calculated as the 12 month percentage change in the CPI. But a different way to think about this calculation is as the average of the 12 one month inflation rates. If there is no change in trend inflation over this period of time, the 12 noise terms will have a mean of roughly zero, and the year-over-year inflation rate will provide a good estimate of trend inflation. The rub is that there have been changes in trend inflation over the past couple of years. Consequently, the year-over-year inflation rate also takes an average of trend inflation over the past year. This is why, in the chart above, year-over-year inflation has consistently exceeded the monthly inflation rate for the last year and a half.

But there’s nothing magic about year-over-year inflation rates, and we’ve previously advocated for annualizing the three month inflation rate as a good compromise between capturing changes in trend in a timely fashion, and seeing through the noise in monthly inflation rates. By construction, the three month inflation rate tracks movements in monthly inflation while removing much of the noise – as can be seen in the chart above. One thing that the three month inflation rate lacks is a fancy name like supercore; oh, and the NYT soap box.

By Paul Gomme and Peter Rupert

The BLS released the Job Openings and Labor Turnover Survey (JOLTS) for May on July 6. The data reveal a pretty mixed view. Job openings fell 496,000 and now stands at 9.8 million vacancies. The number of unemployed workers in May was 6.1 million so that there were 1.6 jobs available for each unemployed worker.

The JOLTS data also contains information on the rate at which workers are hired, laid off, quit and job openings. The rates are determined by dividing by the level of employment. The JOLTS covers about 95% of all nonfarm payroll jobs in the US. While the openings rate declined the hiring rate and the quit rate rose. The layoff rate remains at one of its lowest rates since the inception of the JOLTS in December of 2000. Note that all of the other rates are well above their historic average.

The JOLTS data provides some evidence for those looking for nascent signs of a recession. Quit and hiring rates fell through the March-November 2001 and December 2007-June 2009 recessions, and both series started falling prior to the latter recession. The declining quit and hiring rates since early 2022 fit this pattern. Unfortunately, JOLTS only covers three recessions, so it’s tough to make much of the historic precedents, particularly since the February-April 2020 recession was due to COVID-19. Another potentially confounding factor is that both quits and hires are at historically high rates, and the recent declines may be due to reversion to the mean.

The data on initial and continued claims for unemployment insurance also came out on July 6. Initial claims rose slightly but remain relatively subdued. Continued claims have been trending down over the past few months.

On July 7, the employment situation report for June was released by the BLS and showed a payroll employment increase of 209,000 but downward revisions for April and May totaling 110,000 for the two months. One way to think about these changes is that roughly half of the employment gains reported for June were offset by the downward revisions in the previous two months. The employment change in the private sector was the smallest increase over the past couple of years.

The phrase “little changed” was peppered throughout the press release for the Employment Situation Report. The report highlighted gains in employment in government (60,000), health care (41,000), social assistance (24,000) and construction (23,000). Yet the gains in professional and business services, and leisure and hospitality (both 21,000) were described as “little changed”. It would seem that the difference between a gain warranting notice, and “little changed” is 22,000.

Average hours worked in the non-farm sector inched up 0.1 hours to 34.4. That increase, combined with the increase in private employment led to about a 5% in total hours of work in the private sector. There was little change in average hourly earnings.

One way to combine data from JOLTS and the employment report is through the Beveridge curve which plots the vacancy rate (from JOLTS) against the unemployment rate (the household survey). As seen in the chart below, the data up to and including the Great Recession (2007-09) appears to lie on a stable Beveridge curve. After the Great Recession, the Beveridge curve shifted out. And after the pandemic, the Beveridge curve shifted out yet again, with the most recent data in the north west quadrant. One interpretation of an outward shift in the Beveridge curve is that it reflects lower efficiency in the matching process between jobs and workers. Under this interpretation, at a given level of unemployment, firms need to post more vacancies in order to fill jobs. An alternative interpretation is that the outward shifts in the Beveridge curve are due to lower costs of recruiting workers: firms post more vacancies because doing so is simply much cheaper that in the past. One way to distinguish between these two alternative interpretations is to look at how quickly vacancies are filled: less efficient matching says it should take longer while lower job posting costs implies a shorter period of time. Unfortunately, the speed at which vacancies are filled is no longer available.

According to the BLS household survey, employment increased 273,000, the number of people unemployed fell by 140,000 and the labor force increased by 133,000. The participation rate and employment to population ratio were unchanged. The unemployment rate fell from 3.65% to 3.57%.

By Paul Gomme and Peter Rupert

The April employment report was released by the BLS and revealed a 253,000 increase in payrolls. However, there were downward revisions totaling 149,000 (down 78,000 in February and 71,000 in March) that threw a little cold water on the report. There were few sectors that had any decline except for temporary help services that shed 23,300 jobs. The workweek held steady at 34.4 hours so that total hours worked increased by 2.1%.

Measured over the past year, average hourly earnings rose 4.45% in April, up from 4.3%. However, when measured relative to the previous month, earnings growth accelerated from 3.3% to 5.9%. As the figure below shows, month-to-month growth rates for earnings are quite choppy. The 3 month change is somewhat smoother; measured this way, earnings growth rose from 3.4% to 4.3%.

The household survey showed that the labor force participation rate remained at 62.6% despite the labor force falling 43,000. The employment to population ratio was also unchanged at 60.4%. The unemployment rate fell from 3.50% to 3.39%.

Unemployment insurance claims spiked up to 264,000, the highest since October of 2021. Continued claims, however, were little changed.

Between March and April, there was little change in inflation as measured by 12 month percentage change in either the Consumer Price Index (CPI) or core CPI (excluding food and energy). However, the annualized monthly percentage change in the CPI rose from 0.6% to 4.5% while core CPI rose from 4.7% to 5.0%. As we have stressed in earlier posts, these annualized inflation rates are quite volatile while 12 month percentage changes respond sluggishly to changes in trend. The 3 month annualized percentage changes strike us as a good compromise between smoothing and quickly capturing trend changes. On this basis, CPI inflation was down slightly, from 3.8% in March to 3.2% in April; core CPI inflation was essentially unchanged at 5.1%. All of these measures of CPI inflation are currently running well above the Fed’s target of 2%.

Given that CPI inflation is higher than the Fed’s 2% target, it may not be surprising that inflation expectations similarly exceed this 2% target. While the May readings for the 1 year and 2 year expected inflation are unchanged at 2.65% and 2.4%, respectively, the 5 year and 10 year expectations rose marginally.

It seems that monetary policymakers no longer look at what’s happening to money growth. That the Fed changed its definition of monetary aggregates starting in May 2020 makes it difficult to take a long view on money growth. Nonetheless, since May 2021 (given the change in the definition of “money” in May 2020, the earliest date for which year-over-year growth rates can sensibly be computed) growth of the monetary aggregates M1 and M2 has slowed. Indeed, both have been contracting since late 2022. A traditional monetarist like Milton Friedman would likely look at the chart below and predict future deflation. One way to think through all this is via the quantity theory of money: Mv = PY where M is money, v velocity, P the price level, and Y real output. This relationship can be recast as: money growth + velocity growth = inflation + real output growth. If velocity is roughly constant (so that its growth rate is 0), and long run real output growth is constant, the quantity theory of money predicts a tight relationship between money growth and inflation. As Milton Friedman put it, “Inflation is always and everywhere a monetary phenomenon.”

Of course, there have been important developments within the banking system. One such development is that the Fed now pays interest on excess reserves of banks held at the Federal Reserve Banks (“excess” meaning above-and-beyond what is required to satisfy reserve requirements). Plausibly, changes in the gap between this interest rate on reserves and the Federal Funds Rate (the rate banks pay in an overnight market for reserves) might explain the above deceleration of money growth. However, as shown below, the interest rate on reserves and the Fed Funds Rate move in lock step.

By Paul Gomme and Peter Rupert

The BEA announced that Q1 real GDP increased 1.1% at an annual rate. Many in the press have, obviously, noted the declines over the last two quarters indicates the economy is slowing. Indeed, today’s report “undershot” expectations that were around 2%. However, the underlying components were a bit more mixed and the strength of the economy might be in the eye of the beholder. Real consumption spending increased 3.7%, the largest increase over the last 7 quarters, and disposable personal income rose 8.0%.

The output identity, Y = C + I + G + X – M, tells us the uses of output (the “demand” side). If output growth is down, then one or more of the right-hand side components must be down. From above, consumption growth rose from 1.0% in the fourth quarter to 3.7% in the first quarter. Growth in government spending likewise rose, from 3.8% to 4.7%. However, real investment spending fell 12.5% in the first quarter compared to a 4.5% increase in the fourth quarter. Within real investment, real inventories grew 0.4% (up from -3.7%), non-residential structure investment grew 11.2%, and residential structure investment fell by 4.2% (although this decrease was an improvement over the very negative growth rates for this category late in 2022).

While it has become standard in the U.S. to annualize the quarterly growth in GDP, one could also look at the year-over-year growth rate. Now, Q1 does not look so bad relative to the past few quarters and, in fact, has actually increased over the previous quarter.

The Personal Consumption Expenditure Price Index (PCE) rose 4.2% in the first quarter compared to growing 3.7% in the previous quarter, obviously high above the FOMC’s 2.0% target.

These developments lead to a murkier monetary policy outlook. Output growth has slowed, perhaps reflecting the cumulative effects of monetary policy tightening over the past year. Yet, PCE inflation is still well above the Fed’s 2% target (see also the CPI and PPI), and consumption growth actually accelerated in the first quarter. In previous posts, we have noted that the labor market is still quite strong, with an historically low unemployment rate, and roughly two job openings for every unemployed person. Given that, there is nothing new to dissuade the FOMC from another rate hike or two.

By Paul Gomme and Peter Rupert

According to the Establishment Survey from the Bureau of Labor Statistics , employment increased by 236 thousand in March 2023. While this increase is the smallest thus far for 2023, it’s close to the average for the second half of 2022. The BLS also revised the change in employment for February up from 311 thousand to 326 thousand, and revised its reading for January down from 504 thousand to 472 thousand.

The BLS also reported that even though employment was up, average weekly hours were down, so that total hours worked fell 6.8 million hours in March which follows on the heels of a 4.1 million hour decline in February. The chart below shows considerable variation in the change in hours worked.

Private sector employment was up 189,000 and the government sector added 47,000 jobs. The private service sector added 196,000 jobs while the goods producing sector shed 7,000 workers. The largest gain in the service sector came from Leisure and Hospitality, adding 72,000 jobs.

Commentators have noted the decline in average hourly earnings, down 4.24% compared to March of 2022. This has been portrayed as a positive development for policymakers. The reason for this positive portrayal is that if higher inflation expectations become entrenched, then workers will want higher wages (to compensate for the higher inflation) which may lead to wage-price spiral in which wage increases feed to price increases which, in turn, feed into further wage increases, and so on. However, as with our commentary on inflation, year-over-year wage changes (the ones discussed by commentators) are slow to pick up changes in trend. And, as with annualized monthly inflation rates, month-to-month changes in average hourly earnings can be quite noisy; indeed, the annualized percentage change over the month actually rose 3.3% compared to 2.5% for the previous month. A 3 month average of monthly earnings growth smooths out some of the monthly fluctuations while, at the same time, picking up changes in trend in a timely fashion.

Readers of this blog know that inflation has generally been running higher than earnings growth. In previous posts, we have plotted average hourly earnings along side various price levels. Here, instead, we present real average hourly earnings: earnings divided by measures of the general price level. So that these measures are all comparable, we divide the consumer price index (CPI) by its average for 2012. The choice of 2012 is motivated by the fact the personal consumption expenditures (PCE) price index is already set to equal 100 in 2012. The chart below expresses average hourly earnings in 2012 dollars. Both CPI measures exhibit declines since 2021: For the overall CPI, average hourly earnings have declined by roughly $1 per hour, while core CPI records a decline of $0.42 per hour. Alternatively, using the PCE price index, earnings fell by $0.36 per hour; using the Fed’s favorite price index, core PCE, earnings have been essentially flat since the start of 2021.

The Household Survey from the BLS showed a 577,000 increase in employment. Moreover, the labor force, the participation rate and the employment to population ratio all increased. The unemployment rate fell from 3.57% to 3.50%. The Jobs Opening and Labor Turnover Survey still shows significant job openings even though they have come down some over the last few months.

Overall, the employment report shows a still-strong labor market, albeit there are signs of slowing in some areas. The Fed seems to have calmed inflation at this point and, depending on what unfolds over the next month or two, there may be a pause in raising rates since monetary policy tends to work with a lag.

By Paul Gomme and Peter Rupert

The BEA announced a downward revision from 2.7% to 2.6% in the third and final estimate of Q4 GDP. The revision primarily reflected downward revisions to exports and consumer spending. The final estimate for 2022Q4 GDP did not change the overall stance of the US economy.

Several sources picked up on the fact that the PCE price index, the primary inflation index cited by the Fed, fell. CNBC reported: “On a 12-month basis, core PCE increased 4.6%, a slight deceleration from the level in January.” The New York Times remarked,

The measure of inflation most closely watched by the Federal Reserve slowed substantially in February, an encouraging sign for policymakers as they consider whether to raise interest rates further to slow the economy and bring price increases under control.

The Personal Consumption Expenditures Index cooled to 5 percent on an annual basis in February, down from 5.3 percent in January and slightly lower than economists in a Bloomberg survey had forecast. It was the lowest reading for the measure since September 2021.

NYT, March 31, 2023

However, as we remarked in an earlier post, measured inflation is sensitive to the time horizon over which it is computed. Annualized month-to-month inflation rates are quite volatile; year-over-year inflation is much smoother, but is slow to reflect changes in trend inflation. And it’s trend inflation that policymakers, among others, are concerned about. Yet, identifying trend inflation is difficult. One approach is to average monthly inflation rates over a few months. In the figure below, we include the 3-month average of monthly inflation rates which is our attempt to balance the desire to smooth monthly inflation rates while capturing changes in trend in a timely fashion.

The Fed tends to focus on so-called core PCE inflation for which the volatile food and energy components are removed. Presumably, the reason for looking at core PCE inflation is that it gives a better read on trend inflation. To be sure, core PCE inflation is less volatile, as shown in the chart below. Nonetheless, this measure continues to exhibit large fluctuations when measured on a monthly basis. And, it should not be surprising that the annual (year-over-year) inflation rate is smoother, but slow to reflect changes in trend inflation. Again, we include the 3-month average of monthly inflation rates.

One way to think about the problem of extracting a measure of trend inflation from the data is that observed inflation is composed of trend inflation and an ‘error’ which reflects issues in measuring trend inflation. In the jargon of the profession, we face a signal extraction problem: we are trying to extract the signal (trend inflation) from noisy data (measured inflation). Averaging monthly inflation rates over some horizon (for example, over 3 months, as above) can be thought of as canceling out the positive and negative error terms in this signal extraction problem. (Again, using some jargon, ideally the errors are uncorrelated over time, or independently and identically distributed.)

Finally, what are the policy implications of the latest PCE price index numbers? As mentioned above, some commentators have noted that PCE inflation was down in February relative to January. True enough when looking at either the monthly inflation rate, or the year-over-year inflation rate. However, the 3 month average inflation rate is up. It is hard to make a strong case that there has been a change in trend inflation. In any event, all of the gauges of inflation presented above continue to run well above the Fed’s stated 2% target.