Base effects are back in the economic conversation as inflation falls from its mid-2022 peaks. So what are base effects and why do they matter? Simply put, the term base effect refers to the fact that measures of change over time depend on what prior period you are comparing to. OK, well, duh. But it turns out that this can make a big difference in interpreting real-world statistics. Take inflation. Economists (including me) who report on inflation usually do so on a year-over-year basis, calculating the inflation rate as the amount of price change since the same month a year ago. There is a good reason for doing this, since inflation is very choppy month-to-month, and making comparisons over a longer time frame smoothes out short-term volatility and does not give too much weight to an idiosyncratic monthly jump or drop. But this can also give a biased views of developments during a time period when inflation is rapidly rising or declining. For example, last fall year-over-year inflation measures were comparing price levels that included the summer run-up to the much lower average prices that prevailed at the end of 2021. That produced high measured estimates of year-over-year inflation that only gradually receded even as monthly changes had decelerated considerably. You can see this in the graph below that compares year-over-year consumer price inflation to a measure that compares average prices over the most recent three months to those prevailing in the preceding three-month period. Notice how the CPI path flattened after June, 2022 and how the year-over-year inflation measure didn't fully reflect that slowing until very recently. Calculating something like this, or using a three-month moving (trailing) average, allows us to see recent developments that may not be obvious in the year-over-year figures. In recent months, base effects have tended to pull measured inflation downward. This has been most dramatic for some inflation components that were temporarily expensive during the pandemic and its aftermath. As gasoline and other fuel prices have fallen back from high levels last summer, year-over-year comparisons to those temporarily high prices have resulted in large measured price declines, even though the prices are now pretty stable month-to-month. Same thing for new and used auto prices, which soared during the pandemic when component supplies were curtailed. Food prices may soon do the same thing. So as we get beyond the period of temporarily super-high prices, the end of favorable base effects is one reason that some economists (including Yours Truly) expect that progress bringing inflation down the rest of the way to the Fed's 2% goal will be harder and take longer than the progress we have seen so far.

The manufacturing sector continues to contract. The ISM® Purchasing Managers Index ® (PMI®) for manufacturing has been below 50 for eight month in a row, and in June was at its lowest level since May 2020 during the very beginning of pandemic recovery. (A value below 50 indicates that more firms than not report contracting business activity. I discuss this here .) Manufacturing industrial production has been flat to slightly down since December of last year, after moderate growth in 2022. Weighing on manufacturing have been higher borrowing costs, upward wage pressures, lingering disruptions to supply chains, and weak demand as consumers have shifted toward service consumption after the boom in purchases of physical goods during the pandemic. As a result of weaker production activity, manufacturing employment has softened considerably, falling in two of the past three months, even as the economy overall has continued to add jobs at a (too?) healthy pace. Nearly all sub-sectors of manufacturing have seen a steep downshift in hiring since last fall, with the sharpest effects in nondurable goods, particularly in industrial supplies like chemicals and plastics, as well as paper and printing. The only major category still experiencing robust employment growth is the motor vehicle sector, where demand and production are moving upward as auto sales recover from very low 2022 levels. It would be nice if we could find signs that the manufacturing recession is bottoming out. Unfortunately, more detailed indicators within the ISM® Purchasing Managers Index® paint a pretty pessimistic picture. Notably, forward-looking indicators of current and backlogged orders are among the most negative measures. In particular, a far greater number of firms have been reporting a falling backlog of orders compared with those reporting an expanding order book. (Separate Commerce Department data on new factory orders also show weakness.) If we want to look really hard for positive news in the ISM® Report on Manufactures®, then in addition to a somewhat less widespread decline in new orders in June, inventories have been moving in the "right" direction, with a substantial proportion of firms reporting that their own inventories and those of their clients are contracting, potentially setting the stage for future production. But this is pretty weak evidence that the manufacturing environment is set to improve. And at this point, the recent flatlining of manufacturing production is nothing like the decline in output that we might expect if the overall US economy slows sharply or enters a recession later this year or in 2024. In a typical US downturn, industrial production drops 4-9% from peak to trough (it fell much further during the two very severe recent recessions). So manufacturing activity could decline more significantly if demand drops off in coming months. How large a role might manufacturing itself play in causing a US recession to begin? Perhaps not very much. Manufacturing represents a relatively small 11% share of US GDP and an even smaller 8% of total US employment . What happens in manufacturing is important for these industries and their workers, but it is not a dominant force in the overall US economy.

Today begins the two-day FOMC meeting on interest rates. The Fed is widely expected to raise the federal funds rate by a quarter point to the 5.25-5.50% range. How many further rate hiking might be in store? An essential element of FOMC decision-making has been the state of US labor markets, in particular their stubbornly persistent tightness. (Of course other factors will be important, particularly recent inflation progress, which I'll touch on below.) Before the June FOMC meeting, I reviewed a set of five labor market indicators that might be considered by the Fed. I am going to update them here and make a few brief comments. Here are the five labor market indicators: 1. Wages. Data on new wage offers from Indeed.com 's Indeed Wage Tracker have continued the steady decline we have observed before, now standing at 5.1%. However, there has been no additional progress in easing of the government's measure of average hourly earnings, which have flat-lined at 4.4% for three months. Both measures tend to average about 3% over the long run. Verdict : Wage pressures continue to be stubbornly persistent , even if the Indeed.com data on new wage offers provides hope that further progress may be coming. 2. Claims for unemployment compensation. The number of first-time filers for unemployment benefits picked up in June and July but has fallen back a bit in recent weeks. Verdict : This still looks relatively favorable but not showing any signs of additional labor market softening . 3. Job openings. The number of open jobs relative to the available labor pool has been trending downward, but it remains well above pre-pandemic levels. While some industries have seen continued easing, the overall picture remains largely the same as before the June FOMC meeting: a very slow decline in open jobs. Verdict : An encouraging trend continues, but there has been little overall decline in the past month. Businesses are still having unusual difficulty finding qualified workers. 4. Change in payroll employment. The net number of jobs created by US businesses has been on a downward trend for the past two years but rose in April and May. Since my last report, June data has come in at only 209,000 jobs added, the smallest increase since a decline in December 2020. While the pace of deceleration remains slower than in earlier stages of the recovery, this is heartening news that hiring may be pulling back. Verdict : The pace of hiring has resumed easing , but we are not experiencing the kind of job contraction that we would normally see at or before the start of a recession. 5. Unemployment. The unemployment rate has been running at or near half-century lows. After ticking up to 3.7% in May, it fell back to 3.6% in June. Verdict : The proportion of the labor force looking for work remains unsustainably low . Updated verdict for the labor market and this week's Fed decision In my mind, there has been no appreciable change over the past month in labor market developments, other than relief from May's scary job surge. There has not been enough additional softening to make much of a difference for tomorrow's FOMC decision. The labor market remains tight and therefore likely to continue to drive inflation that exceeds desirable long-run performance. But, as I mentioned before, by many measures progress toward easing labor market pressures is evident and ongoing, even if glacial. The economy is on track to slowly converge to satisfactory long-run labor market conditions. It is not clear that the Fed needs to raise interest rates further to accomplish this, but then I am not on the committee! What else will the FOMC be looking for? While Fed Chairman Powell has repeatedly put emphasis on the tight labor market as a key consideration in setting interest rate policy, that is not the only factor that will weigh on this week's decision, and perhaps more importantly the possibility of one or more hikes later this year. Perhaps the most promising development since last month was the June CPI report , which showed continued inflation deceleration, to roughly 3%. And the Fed's bugaboo SuperCore inflation (services less housing) has continued to decline at an encouraging rate within the CPI data. BUT , the Fed has tended to emphasize the Personal Consumption Expenditure (PCE) deflator, which has shown almost no letup in its SuperCore component . New data on the PCE won't be available until the end of the week. Given Chairman Powell's past statements, the overall labor market picture is consistent with the anticipated quarter-point FOMC hike this week. What will be more interesting will be to hear what he has to say about ongoing labor market developments and what they might mean for future Fed actions.

Out today is the Institute for Supply Management's® Services Purchasing Managers Index (PMI ®). A good opportunity to discuss diffusion indexes. A diffusion index reports the difference between positive and negative responses, in this case, the difference between those purchasing managers for businesses who are reporting expanding activity and those who are reporting contracting activity. The ISM® reports these both for the manufacturing sector and the services sector. As the Institute for Supply Management® reports the index, values above 50 mean that more than half of individuals surveyed report expanding activity; readings below 50 indicate that more than half of respondents report contracting activity. How far away from 50 you are does not indicate how severely activity is expanding or contracting, but rather how widespread expansion or contraction is. As you can see, the manufacturing PMI® has been contracting now for seven consecutive months, one indicator that the manufacturing part of the economy is already in recession. The service index has generally been above 50, although it dipped below 50 in December and was just barely above in May. By the PMI®, services sectors of the economy have been doing much better than manufacturing, but they still show signs of weakness. The ISM® also reports various sub-indexes that give more detailed info. You can read more about it on their web site .

Today saw the release of the January JOLTS data by the US Bureau of Labor Statistics. JOLTS (Job Openings and Labor Turnover Summary) data track, well, just that. How many job openings there were in January, and information on turnover: the number of hires and separations from employment for various reasons. These days we are tracking the JOLTS for indications of how tight the labor market is and where it might be headed. The quick take on January's report is that the number of job openings has fallen back just a tad, but remains high. There were slightly more layoffs than in December, but they remain pretty low. The Big Picture hasn't changed from recent months: There are a VERY large number of job openings these days, one indicator of how tight the US labor market is. Here is a graph of the difference between employment and open jobs on the one hand, and the size of the labor force on the other. This is a meaningful measure of excess demand for labor, since it tells us how many bodies firms are willing to employ compared with the number of people working or available for work. In January there were about 5 million more jobs than workers. There is considerable excess demand for labor compared with the pre-pandemic period, and that this has not really fallen off at all. The labor market is still very hot! Part of this is due to strong business demand for workers, but part is also due to the failure of the labor force to recover to its previous level. But that's a story for another day... Job openings data is available by industry, and taking a look at just a few, we can see that there are higher numbers of open jobs across the board compared with the number before the pandemic. Employees are particularly hard to come by in manufacturing, where the number of openings is about twice what it was in December 2019. (BTW, what I have done in the graph is to "normalize" each industry's job openings to a value of one in December, 2019, so we can compare today's numbers relative to a common starting point. Also, because the data jumps around a lot from month to month--we say it is "noisy"--I am showing the average value over the past three months.) These days the Federal Reserve is particularly concerned about the state of the labor market, because of upward wage pressure that tends to come when labor is in excess demand. The JOLTS data is just one of the federal datasets that can shed light on the state of the labor market. February data on jobs, employment, and unemployment is coming soon.

On Sunday, OPEC made a surprise announcement that it would reduce crude oil production by 1 million barrels (bbl) per day, rising to perhaps 1.6 million bbl/day by late summer. Oil futures have jumped to about $80 recently in anticipation of the effect of the production cuts on the global oil supply-demand balance. Some analysts have predicted $100/bbl oil in coming months as a result of the production cuts. Certainly the cut in OPEC production is unwelcome at a time of rising stress on the US and global economies. Higher crude oil prices will raise business costs and lead to a temporary upward blip in inflation. If that feeds through into ongoing inflation, it will make the Fed's disinflation job harder, potentially leading to higher interests for longer. But a little perspective is important. In inflation-adjusted terms, even at $80/bbl crude oil prices are running below the average price that has prevailed over the 2004-2022 period. (Prices averaged about $88.50 over this period, measured in 2022 dollars.) A price of $100 in June would equal about $94/bbl measured in 2022 dollars.* That would be nearly a third higher than the March average price, but only 7% higher than the long-run average. At that level, it is hard to see how this would have a significant adverse impact on business costs and the macroeconomy. To be sure, no one knows how high oil prices will actually move in response to the OPEC supply cut. A higher peak would have a bigger adverse macroeconomic effect. It would have been nice if they had held off for now, but spending needs of some OPEC governments apparently drove the decision. One clear adverse consequence is that, despite the Western embargo on Russian oil, higher world market prices will help fuel the Russian war machine. Oil prices have implications that extend beyond their purely economic ones. *I have to assume something about US consumer price inflation in coming months in order to calculate the future real price of oil measured in 2022$, if the nominal market price rises to $100 by June. I assume the CPI slowed from 6% in February to 5.5% in March, but then accelerates back to 6% for the April-June period because of the higher oil prices. Different reasonable assumptions would not materially affect the real price estimate.

The Federal Open Market Committee raised the federal funds interest rate to an upper bound of 5% today, following their March meeting. Their press release indicates ongoing concerns that robust spending, tight labor markets, and still-high inflation justify the rate hike. While the Fed acknowledged concerns that recent banking problems are likely to result in somewhat tighter lending conditions, they clearly believe that inflation trumps these concerns--at least for now. The Fed's two-pronged approach to the twin ills of inflation and banking sector security seem appropriate. Over the past week, the central bank has accommodated the extra cash needs of financial institutions through a special program for Silicon Valley Bank and other directly affected institutions, and it has permitted a high volume of general financial institution borrowing at the discount window. According to the AP , the Fed lent about $300 billion to financial institutions over the past week, more than half of that at the discount window. This meets the Fed's need to provide liquidity in times of crisis. Today's interest hike need not be seen as inconsistent with this goal. While it is true that the financial turmoil has likely raised the cost of finance for some borrowers, estimates of the effect are all over the place . (Chairman Powell suggested this is perhaps equivalent to a rate increase or more.) Today's quarter point rise sends the signal that inflation is still job one, but the increase is small enough not to be destabilizing. We and the Fed will be watching real economic activity and banking stability closely before the next FOMC meeting, scheduled for May 3.

Fed Chairman Jerome Powell gave his semi-annual report to Congress today. He was clear that further rate hikes are coming, given recent data that indicates persistent inflation in services and renewed concerns about the ultra-tight labor market. Financial markets reacted by raising their expectations of where the federal funds interest rate will peak this summer, at 5.75%. Source: https://www.cmegroup.com/markets/interest-rates/cme-fedwatch-tool.html You might ask "so what"? Well, these revised expectations of coming Fed policy caused financial investors to bid up short-term interest rates substantially later in the trading day, and in fact to a point where short rates were much higher than long-term interest rates. Such a "yield curve inversion" is often seen as an indicator of a pending recession . We get another jobs report on Friday and CPI for February next Tuesday, which may tell us more about whether the recent upward pressure on prices is likely to continue. The Fed has been raising rates at the most rapid pace since the early 1980s to slow demand and therefore take pressure off prices. The biggest effects so far have been on housing markets and manufacturing, with consumer spending still expanding at a moderate rate. The next meeting of the Fed's policymaking committee is March 22. bg 2/7/2023

Continuing in the “pick a survey, any survey” vein, labor market data out today are playing different tunes, depending on what you are listening for. Different strokes for different folks The Bureau of Labor Statistics' monthly labor report includes results from two labor market surveys for May. The first, the establishment survey, showed robust growth in filled jobs, up 339,000 from April, nearly twice the consensus among forecasters. This strong job growth suggests that at least for now businesses overall are happy to keep adding workers to their payrolls. Strength was not uniform across industries, with some service sectors adding workers at a rapid clip, softer conditions in construction, and net job losses in manufacturing, an area that has been essentially in recession for a half year now. BUT, results of the separate survey of households were more downbeat. The number of people saying that they were employed fell by 310,000. With the size of the labor force holding steady, that pushed the unemployment rate up from 3.4% in April to 3.7% in May. Should I stay or should I go now? The two surveys—one of businesses and the other of households—tend to track each other fairly closely over the medium-term, but they can diverge in the short run. Not only do they come from different sources, but the nature of the data they cover is somewhat different. The establishment survey only covers payroll employees of businesses, while the household survey includes self-employed people. That turned out to be an important difference last month. In the household survey, business employee numbers rose, but the level of “unincorporated self-employment”—which includes independent contractors and some small businesses—fell sharply. It’s hard to know what to make of this. Is the drop in this category a leading indicator of softening activity in personal service areas? Or are we just seeing a readjustment as people are able to pivot away from pandemic-era gig work to "regular" business employment? There is not sufficient detail in the survey to tease that out. Other data in the labor report reinforce the notion that labor markets are not as tight as the payroll employment numbers suggest. Importantly, wage growth continued to ease slightly, although it still remains at a high level that is inconsistent with a balanced labor market and medium-term price stability. What’s the Fed gonna do, gonna do with it? Today’s labor market report one of the last major pieces of data before the Federal Open Market Committee meets June 13-14. (The May CPI will come out the day of the meeting.) The question is whether the gangbusters payroll jobs data will be sufficient to prompt another quarter-point interest rate hike. One might think so, since other recent indicators—like PCE inflation data—have also come in strong. But comments by Fed officials have broadly hinted at a rate-hike pause in June to see how the past year’s monster rate increases are feeding through to the broader economy. And the Fed these days seems to want to avoid surprising markets if it can. So a June pause may be baked in to Fed thinking. But don’t be surprised if the Fed is back to playing more ominous music in July.