Central banks around the world have one thing in common and that is their mandate of keeping prices stable for all consumers. The majority of the world population would be unable to tell you what exactly a central bank does or why they take certain actions, but it is widely known that if inflation increases, one institution is the main culprit: the central bank. As a consequence, there is tremendous pressure from across the spectrum of citizens, whether it be government officials or retirees, they will all be demanding action from central banks to keep prices stable. Significantly slowing the labor market is the most efficient and dependable path to curtailing economic growth and subsequently inflation.
It is always best to start the process of data analytics for economics by looking at leading indicators. Using this process, we gain an idea of where the economy is headed in 6 or more months and can be attentive for when coincident data will confirm or contradict the leading indicators. The United States labor market continues its secular strength amid one of the most anticipated recessions in history. Leading economic indicators have been deteriorating for the better part of a year. However, when viewing coincident to lagging “hard” data there is no confirmation of weakness. As such we will analyze the key data points of the labor market and where they currently stand. Finally, using this information I will provide a framework for what investors need to watch before a downturn in the labor market occurs and subsequently how to position their portfolios ahead of time.
Leading Indicators Are Trending Lower
Per my analysis, individual survey data for the labor market has a fairly low correlation to the unemployment rate even on a lagged basis. For example, whether it’s the employment survey from the Philadelphia or Dallas Federal Reserve their correlation to unemployment is near zero. However, when aggregating regional data together into a composite measurement and then attempting the linear regression once more, the correlation jumps significantly higher to 16%. The one drawdown to this method is the relative low amount of data the composite measure contains, only beginning in 2004. It doesn’t provide the opportunity to analyze how the time series interacts during different economic cycles. As is the case with many leading indicators they need to be taken with a grain of salt.
The regional employment composite has been steadily trending lower since its peak in December 2021 and has returned to its long run mean. I believe we can interpret this as the labor market on a leading indicator basis is no longer overheating. This is of course positive news as the monetary tightening from the Federal Reserve has started to show its effect on the labor market. However, considering monetary policy works with a lag of 12-18 months the full force of tightening won’t be experienced in the labor market and subsequently show in “hard” data until March 2023 at the earliest. Leading to my assessment that once the composite indicator dips below its mean and starts to head negative the effect of tightening will soon begin to show in the coincident and lagged data for the labor market. With that said regional data for employment is currently in the goldilocks phase i.e., it’s not overheating or crashing. In my opinion the data won’t stay at this level for long but will continue its trend lower and likely at an accelerated pace as the full force of monetary policy tightening comes through the pipes to the real economy.
Has Coincident Data Broken Down Yet?
Coincident data unsurprisingly is more dependable on gauging the strength of the labor market. This is due to two factors that are very intuitive. First the data is representative of the number of jobs added in an economy such as total nonfarm payrolls. Now if a job is created in the economy this means an individual has more purchasing power than they did previously. According to the Bureau of Labor Statistics the average American has an income after taxes of $78,000 with $67,000 of that amount being disbursed to a diverse array of expenses back into the real economy. Leading to more growth and jobs being created to meet the increased demand. Clearly if jobs are increasing the unemployment rate will be decreasing and vice versa. For those reasons and the fact both time series date back to the 1940s and 1950s respectively, they are two of my favorite indicators to watch for the labor market.
Total Nonfarm Payrolls have been the outperforming statistic for the labor market during the previous twelve months. It has surpassed expectations in the last nine consecutive months. During the month of December 223,000 jobs were added to the economy. Bewildering economists, and portfolio managers alike, who were expecting a rapid deterioration in the labor market soon after the hiking cycle began. These numbers are consistently beating the forecasts and the average amount of jobs added during the history of the time series. In my analysis, the average amount of jobs added every month in America since the 1940s is 123,000, while currently the labor market has been adding above this amount every month since March 2021. Granted the pandemic obfuscates the data to a degree as most jobs in 2021 weren’t created per se but regained as the economy reopened. To highlight how far off the labor market is from recessionary data, nonfarm payrolls on average during a recession lose 386,000 jobs. There is a long way before the labor market deteriorates to recessionary levels and in my opinion, it will take six months or more before we accumulate enough job losses.
One indicator that provides a glimpse into both the labor market and the strength of manufacturing demand is Average Overtime Hours worked. When viewing Average Overtime Hours, a higher number of means that there is consumer strength and companies require more time investment from workers to match the production needs of the economy. We can interpret a declining trend as excess demand from consumers for manufactured goods is drying up and manufacturer demand for labor will likely decrease in the future. When examining recent data, Average Overtime Hours has been declining since its peak of 4.4 hours in February 2022. The decline is not surprising as the economy has been gradually slowing since 2021. But the interesting fact is during a recession Average Overtime Hours declines on average -0.64 hours while it has already declined -0.8 since its peak. Considering other measures of recessions that the NBER closely follows have not deviated to recessionary levels and the high level of economic growth from which we started after the reopening, in my opinion, there will need to be a bigger downside variance to account for these factors before the economy enters a recession. Resulting in a longer and higher contraction in both the amount of overtime hours worked and consumer balance sheets. This is a harsh reflexive effect where the economy is contracting leading to employees working less overtime and earning less money leading to them spending less money and again the economy contracting more.
We have analyzed leading employment data and two measures of coincident data that are providing contrasting perspectives. Where leading but less dependable regional data and paired with Average Overtime Hours worked having already been through recessionary contractions. However, the indicator with the highest correlation to the unemployment rate, Nonfarm Payrolls is showing secular strength and beating forecasts for nine consecutive months. In my opinion this has been the source of economists and investors having misguided forecasts for the labor market during 2022. In order for investors to make an informed decision regarding portfolio allocations we will analyze two more labor market indicators to build a comprehensible understanding of current labor market conditions and what is most probable moving forward.
Outlook For The Labor Market
The highest frequency data we have regarding employment is Initial Jobless Claims which is provided on a weekly basis for the past 55 years. Frequent releases of data in such a manner allow investors and policy makers to keep a closer eye on the economy and subsequently the biggest factor affecting it, the consumers. Data released for Initial Jobless Claims on January 19 show it declined by 15,000 on a week over week basis. The time series has been trending lower for six consecutive weeks. The low for our current economic cycle is 166,000 Claims during March 2022 and in my opinion, with its momentum the last thing equity bulls will want to see is a test of said lows. If that were to happen it would signal to Powell and the Federal Reserve, that they can keep tightening financial conditions potentially even more aggressively as consumers are stronger than anticipated. Nonetheless, Initial Jobless Claims have only been at a corresponding low level three other times since 1967 highlighting the secular strength exhibited but I also believe the low probability the labor market will get much better from its current level moving forward.
Our final indicator we will be analyzing is OECD Manufacturing Employment. This is another time series with a long history that has endured many economic cycles. Data released in December showed a slight increase to 2.60 but since October 2020 the time series has been oscillating on the 0 bound. In my opinion it is indicative of manufacturing company’s skeptical attitude towards the future potential of economic growth. Per my analysis based on month over month, rate of change OECD Manufacturing Employment has the highest correlation to the unemployment rate when there is no lag between the two data points. Resulting in another indicator investors can watch that has a long and dependable history in order to better gauge the labor market. I believe the expedient tightening the Federal Reserve has conducted in 2022 and their current path of holding interest rates above 5% for 2023, will restrict employment and specifically the manufacturing sector from growing. Leading to a higher tail risk to the downside that investors should have their portfolios prepared for.
Takeaways for Investors
In my opinion investors would be shrewd to be attentive to every week’s release of initial Jobless Claims. The high frequency release allows investors to keep their finger on the pulse of the labor market. If a trend of more than a few weeks develops with a delta of +15,000 or above and this is confirmed with Nonfarm Payrolls that are worse than expectations, then I believe there is a high probability the labor market will deteriorate promptly, and portfolios need to be rebalanced to protect capital.
After evaluating five different indicators for the labor market, I believe investors are better prepared to position portfolios taking the stated information into consideration. In my previous paper I recommended that investors turn to alternative strategies to outperform the market in 2023. However, due to leverage constraints and the complexity of implementing such strategies I would propose that investors increase their allocation from growth to defense equities. Per my analysis during a recession the average return of the S&P 500 is -29% compared to a holding some of the biggest defense stocks such as Lockheed Martin, Northrop Grumman, and General Dynamics of +6.19%. The geopolitical disorder that has escalated in Eastern Europe and tensions mounting in American Chinese relationships will act as a tailwind to this allocation. A caveat to be aware of is Lockheed Martin is the main driver of this outperformance with an average return during recessions of +20%. Nevertheless, I believe diversification is always better than having your eggs in one basket and I would personally never allocate all my capital to one company.