Viewpoints

I don’t have access to the Blue Chip survey of economists’ forecasts of various economic data anymore, so I can’t answer my question. But I do have access to consumers’ inflation forecasts and these forecasts are terrible. Plotted in the chart below are monthly observations of consumers’ forecasts of year-ahead inflation as reported in the University of Michigan Consumer Sentiment Survey (the blue bars). Also plotted in the chart are monthly observations of the actual (until revised) year-over-year percent changes in the All-Items Consumer Price Index (the red line). The CPI percent changes are lagged such that they line up with month in which the consumers’ forecasts were surveyed. For example, in May 2020, consumers were forecasting that the year-over-year inflation rate in May 2021 would be 3.2% (the height of the blue bar in May 2020). As luck would have it, the actual CPI inflation rate turned out to be 4.9% (the height of the red line in May 2020). In October 2022, consumers were forecasting that the year-over-year inflation rate in October 2023 would be 5.0%. In fact, it turned out to be 3.2%. In the latest November survey, consumers are forecasting that inflation will be 4.5% in the 12 months ahead. Given that the sum of the monetary base plus commercial bank credit grew by only 0.7% in the 12 months ended October 2023, my bet is that the year-over-year percent change in the CPI in November 2024 will be much lower than 5.0%.

The Federal Reserve Bank of Philadelphia’s state coincident indexes in October were quite mixed, with the balance tilting toward weakness. A full 32 states show declines from September, with West Virginia’s reading down by 1 percent and Montana’s and Mississippi’s indexes falling more than .5 percent. Of the 18 states with increases, the largest was Nevada’s fairly moderate .33 percent. Over the 3 months ending in October, 16 states had declines, with West Virginia off 2.7 percent, and Montana and Mississippi dropping more than 1 percent. South Carolina and Maryland both increased roughly 1.3 percent over this period, which is not an especially large gain for states at the top Over the last 12 months Maryland had an impressive 7.4 percent increase, and Massachusetts and Vermont were up more than 6 percent. 3 states had increases of less than 1 percent, with New Jersey again at the bottom with a .2 percent reading.

The independently estimated national figures of growth over the last 3 months (.5 percent) and 12 months (3.0 percent) both look to be roughly in line with what the state figures suggest.

Part un was written by me way back on March 14, 2020. I should have paid more attention to my 2020 commentary so that I would not have thought that household spending would be less resilient as it has been so far in 2023. Moreover, I would not have called for a recession to commence in Q2:2023.

In Chart 1 below are plotted monthly observations of the M2 money supply as a percent of nominal Disposable Personal Income (DPI). From January 2015 through December 2019, the median value of this ratio was 91.8%. Then, after the federal government started writing Covid-aid checks to households and businesses, checks financed by the Fed and banking system, the ratio of M2 to DPI reached a high of 118.9% in January 2022. As of September 2023, the ratio had declined to 102.1%, much below its January 2022 high, but also materially above its 2015-2019 median value.

State labor markets were soft in October. Florida’s 28,400 increase (.3 percent) was the only statistically significant change in payrolls (California’s 40,200 gain was not seen as statistically significant). The number of states reporting point declines was comparable to the number reporting increases. The sum of the states’ payroll changes was only 43,800 (noticeably smaller than the national 150,000), the lowest such figure since December 2020.

26 states had statistically significant increases in their unemployment rates in October, though none were larger than .2 percentage point. Nevada continued to have the nation’s highest rate, at 5.4 percent while DC was at 5.0 percent. No other state had a rate more than 1 percentage point above the nation’s 3.9 percent, though California, Illinois, and New Jersey were higher than 4.5 percent. Alabama, Florida, Hawaii, Kansas, Maine, Maryland, Massachusetts, Montana, Nebraska, New Hampshire, North Dakota, Rhode Island, South Carolina, South Dakota, Utah, Vermont, Virginia, and Wyoming, all had rates at or below 2.9 percent, with Maryland at 1.7 percent.

Puerto Rico’s unemployment rate fell t0 5.8 percent, and the island’s payrolls increased 2,800.

Monetary policy influences nominal spending in the economy. In the third quarter, nominal GDP grew 8.6% annualized. So far, in 2023, nominal GDP is running at an annualized pace of 6%. That follows a 10.6% gain in 2021 and a 9.1% gain in 2022. The three-year increase, 2021 to 2023, represents the fastest three-year advance in nominal GDP since the mid-1980s.

The economy's nominal growth performance has two critical messages/implications for policymakers and analysts/portfolio managers regarding Fed policy and market rates.

First, except for the non-economic slowdown following the pandemic, it has taken a Fed funds rate equal to or above the growth in nominal GDP to engineer a sustained growth slowdown/recession. The target on the Fed Funds rate is still 75 basis points below the growth in nominal GDP.

Second, many analysts and portfolio managers still expect a return soon to the interest rate pattern of 2008 to 2020. Yet, that interest rate pattern was abnormal, as was the nominal growth path in the economy. Only once did nominal GDP grow more than 5% during those twelve years, which occurred in 2018. The average gain was about 4%.

The interest rate pattern more applicable to the economy's current growth performance and policymakers' intent to lower inflation is from the mid-1980s to the mid-1990s. At the start of that period, the Fed funds rate, as did the 10-year Treasury yield, exceeded the Nominal GDP growth. Then, in the later part, nominal growth and nominal rates were more in line with one another.

The longer it takes the Fed to adjust policy to the current growth dynamics, the longer it will be before the economy slows and market rates fall.

The Federal Reserve Bank of Philadelphia’s state coincident indexes in September were again soft, and the results were very similar to August’s. 19 states show declines from August, with West Virginia down nearly 1 percent. The largest increase was .65 percent in Maryland.in the rate of growth. Over the 3 months ending in September 10 states had declines, with West Virginia’s -2.2 percent the largest, while Montana was also down more than 1 percent. Maryland had the largest gain, at 2.25 percent. The results are less lackluster at the 12-month horizon, with Maryland up 7.8 percent and Massachusetts and Vermont both up more than 6 percent.. 3 states had increases of less than 1 percent, with New Jersey up a mere .20 percent.

The independently estimated national figures of growth over the last 3 months (.7 percent) and 12 months (3.1 percent) both look to be roughly in line with what the state figures suggest.

State labor markets were soft-to-mixed in September. Only 6 states saw statistically significant increases in payroll, though none reported a statistically significance drop. Texas gained 61,400 jobs and South Dakota reported a .9 percent increase. September was the third straight month in which the sum of state job changes fell short of the national figure; the sum of the differences since June is around 150,000.

A full 16 states had statistically significant increases in unemployment from August to September, with a .3 percentage point increase in Illinois. Once again, Nevada continues to have the highest rate of any state in the nation, at 5.4 percent. Nevada and DC had unemployment rates at least one point higher than the national average of 3.8 percent. (California is the only other state with a rate statistically higher than the nation. Alabama, Florida, Hawaii, Kansas, Maine, Maryland, Massachusetts, Montana, Nebraska, New Hampshire, North Dakota, Rhode Island, Utah, Vermont, and Virginia, were all at least a point lower, with Maryland at 1.6 percent.

Puerto Rico’s unemployment rate moved down to 6.0 percent. The island added 2,000 jobs, and the private-sector total set a new all-time high.

Critics argue that the current inflation rate is much lower than the published rate. That is true; based on the current methodology, "real-time" consumer inflation is less than the published rate. Thirty-five percent of the prices used to estimate the consumer price index are not for the current month but reflect the prices over several months, according to the Bureau of Labor Statistics. Most of that involves the owner's rent index. Because rents change infrequently, the Bureau of Labor Statistics measures these service prices over six-month spans.

The owner's rent index is the brainchild of government statisticians and academia, supported by politicians who want a lower, less volatile price index. Yet, the owner's rent index has two "major" problems.

First, it is not a current price. Rent changes, up or down, would not be captured in the CPI until six months after they occurred. So, given its massive weight in the index, owners' rent would result in reported inflation running below current inflation at the beginning of the rent price cycle and overstate it at the end.

Second, it is not an actual price. The owner's rent index is supposed to measure or capture inflation experienced by people who own a house. But no homeowner pays that price. Economists and policymakers often talk about demand destruction from higher inflation, but there is no demand destruction here since no one pays the price.

A "real-time" and more accurate measure of inflation would require a shift back to the inflation methodology pre-1983. That would include house prices and mortgage rates, creating a real-time, more volatile, and higher published inflation rate. So pick your poison---a flawed index or a higher volatile index? Complaints about the current CPI are frequent, but they would only get louder if a shift to a real-time measure of inflation occurred.