Viewpoints: May 2023

The Federal Reserve Bank of Philadelphia’s state coincident indexes in April report gains in all states from March. Vermont had the largest increase (1.27 percent) while neighboring Massachusetts was the other state with a rise of more than 1 percent. West Virginia was, once again, the leading state at the three-month horizon, up a bit more than 3 percent, while Vermont—perhaps the second hilliest Eastern state, after West Virginia—was second. The majority of states again had increases of at least 1 percent since January; Alaska was again the only state with a decline over this horizon. Over the past 12 months, Massachusetts was the leader, with an increase of almost 4 ½ percent, barely edging Texas and New Mexico. Kansas was the only state with an increase of less than 1 percent since April 2022.

The independently estimated national figures of growth over the last 3 months (.74 percent) appears to be short of what the state figures suggest, while the corresponding 12-month result (3.72 percent) looks more or less in line with the state numbers.

Marked changes in state payrolls were limited in April. 5 states saw statistically significant increases, with California up by 67,000 and Indiana seeing a .5 percent increase. Rhode Island had a sharp .8 percent decline, and a few other states (and DC) had insignificant decreases.

A full 14 states had statistically significant drops in unemployment from March to April. Oregon’s .4 percentage point decline (coming on the heels of a .3 percentage point drop in March) was the largest. Nevada continues to have the highest unemployment rate in the nation, at 5.4 percent. California and DC are the only other places with unemployment more than a point higher than the national averages of 3.4 percent. Alabama, Montana, Nebraska, New Hampshire, both Dakotas, and Utah are more than point under the national figures, with South Dakota remaining at 1.9 percent. Indeed, 17 states, including Florida have unemployment rates below 3 percent. On the flip side, along with California’s 4.5 percent rate, New York and Texas both have jobless figures of 4 percent.

Puerto Rico’s unemployment rate remained at 6.0 percent, but its job count moved above 950,000 for the first time since 2009. Payrolls on the island had (excepting the Maria and COVID shocks) been under 900,000, but have been on a steady increase in this expansion. The record high was October 2004’s 1,059,200. However, Puerto Rico’s job markets has shifted dramatically since then: private payrolls are now only 400 shy of their peak.

The wage cycle is a critical factor in the scale and length of the Fed tightening cycle. Based on the current wage data, history says the tightening cycle has yet to reach its peak rate, and the duration of the higher official rate cycle could extend much farther than the markets expect.

The thinking behind the Fed hiking rates to break the inflation cycle is straightforward: lift rates to a prohibitive high enough level that curtails or breaks the willingness to borrow and spend. Each tightening cycle is different, and the scale and length often depend on wage and income growth.

One traditional way to determine if higher rates are prohibitively high is to compare them to inflation. That helps determine the real borrowing costs for businesses since the price is what firms get for their products and services. Yet, to measure the real borrowing costs for consumers, one needs to compare interest rates to wages since the latter is the worker's price.

In April, and for the first time since the Fed started to raise official rates in March 2022, the gap between Fed funds and wage growth was closed. That's the good news. The bad news is that the tightening cycles of the late 80s, 90s, and mid-2000s ended when official rates were several hundred basis points over the wage growth. So history would say the Fed tightening cycle is far from over, and the April wage and jobs data lends credence to that view.

Still, policymakers may pause and gauge the lagged effects from the scale of the tightening to date. Lagged effects from monetary tightening are adverse and build over time. Still, the overall stance of monetary policy must be tight or restrictive for them to generate the negative economic and financial results policymakers want to achieve.

Up to this point, the policy stance shifted from less accommodative to neutral. That helps to explain why cyclical sectors (motor vehicle sales in April were the highest in nearly two years, and housing activity has perked up) showed renewed momentum. More rate hikes will be needed to break the momentum in cyclical industries.

In Q1, the combined output of the cyclically sensitive motor vehicles and residential housing sectors expanded by 1.3% annualized, slightly better than the 1.1% growth for the overall economy and the first quarterly gain since late 2021. Also, the Q1 data shows that operating profits gained sequentially quarter over quarter and year over year. The rebound in cyclically sensitive sectors and profit data run counter to the recession forecasts. All economic recessions have standard features; declines in cyclically-sensitive sectors and drops in operating profits. Those features are missing at this time.

S&P purchasing managers manufacturing index rose over one percentage point to 50.2 in April. That’s the highest level in six months, driven by new orders, production, and employment gains. Thus, the rebound in cyclically sensitive sectors has continued into Q2.

Recessions forecasts are linked primarily to the inverted yield curve and the decline in the leading indicators. Questions over the accuracy of the signal from the inverted curve stem from the Fed's new policy tool, quantitative easing (QE). Since the Fed now actively purchases substantial quantities of long-duration fixed assets to keep a lid, or even depressing, on long-term interest rates, how can the yield curve signal be as reliable as in prior periods?

History shows that lower long-term borrowing costs often lead to faster growth in cyclically-sensitive sectors. The yield on the 10-year Treasury has declined 75 basis points in the past six months, and cyclically sensitive sectors have rebounded. Is that a coincidence, or are they interrelated? If the latter, the recessionary signal from the inverted yield curve is wrong. It’s the latter.

The leading economic index, which has declined sharply over the past year, triggering fears of recession, includes the yield curve. Yield curve inversion has been a significant factor in the decline of the aggregate index over the past year. Yet, is the yield curve still a reliable leading indicator with the creation of QE?

It’s common for the index composition to change from one cycle to the next because economic, financial, or policy changes make some indicators less reliable or obsolete. Broad money failed as an indicator before the Great Financial Recession. A new credit series replaced it in 2012. It will not be surprising if the leading index includes a QE series and removes the yield curve indicator at some point.

It’s worth noting the 2020 recession was unique from the standpoint non-economic factors triggered it. Yet, the monetary and fiscal policymakers viewed it as a vast economic disaster, rightly so, and responded with the most significant monetary and fiscal stimulus ever seen. Doubling the Fed's balance sheet from $4 trillion to over $8 trillion in 18 months was never done before, and we still need to learn all the economic and financial consequences. At the very least, the aggregate stimulus and new ways of interjecting liquidity in the system raise questions over long-trusted indicators such as the yield curve and broad money.

Investors should keep it simple; the economy is growing if companies generate profits and hire.

The Bureau of Economic Analysis’s first guess at real GDP annualized growth in Q1:2023 was a paltry 1.1%. Bear in mind that the BEA does not yet have March 2023 data for business inventories, net exports or construction expenditures. With only full January and February inventories data, the estimated change in real business inventories in Q1 “contributed” minus 2.3% to the annualized percent change in Q1 real GDP. Who knows, perhaps the March inventories data will reduce the magnitude of the drag on real GDP growth from this component.

In contrast to the drag on Q1 real GDP growth from real business inventories, real Personal Consumption Expenditures (PCE) contributed 2.5% to Q1 annualized real GDP growth. At an annualized rate, Q1 real PCE increased 3.7% versus 1.0% in Q4:2022 and the fastest growth in real PCE since the 12.1% recorded in Q2:2021. Given that real PCE accounted for around 71% of real GDP in Q1, the economy seemingly was on fire in Q1, right?

Wrong! Let’s take a look at the behavior of monthly real PCE as compared to its quarterly average. This is shown in Chart 1 below. The blue bars in Chart 1 represent the month-to-month annualized percent changes in real PCE. The red bars represent the quarter-to-quarter annualized percent changes in real PCE. So, the annualized growth of 3.7% in real PCE in Q1:2023 was the result of the outsized 17.6% annualized growth in January 2023 PCE growth. Monthly real PCE contracted in February and March 2023. In fact, real PCE contracted in four of the past five months. It seemed as though a myriad of measures of economic activity were very strong in January, very strong, as a former president might say. Could it have been that January 2023 was uncharacteristically warm?