Why The Unemployment Rate Is Higher Than Reported

Summary

• Markets rocketed higher on Friday due to a much better-than-expected May unemployment figure.
• Not mentioned by most of the media, there was a substantial rise in likely incorrectly filled surveys and a decline in survey response rates.
• Accounting for these people, the "true" unemployment rate is actually likely closer to 17%.
• Including the drop in the labor participation rate since February and those employed (but not producing) due to temporary PPP loans, the adjusted figure is likely over 30%.
• This seemingly "good" economic news increases the odds government stimulus (in which the market depends) will be ended prematurely.
• Looking for a helping hand in the market? Members of Core-Satellite Dossier get exclusive ideas and guidance to navigate any climate. Get started today »

(Pexels)

Despite my bearish view, markets rocketed higher on Friday as the May employment numbers were far better than expected. Wall Street expected a 19.5% unemployment rate, significantly higher than April's 14.7% rate. However, unemployment actually declined by 2.5M to a rate of 13.3% in May.

This has caused large gains for equities with the S&P 500 (SPY) rising about 2% on the announcement, leaving the index nearly positive year-to-date. The news also caused further declines in long-term bonds with the 20-year government bond (TLT) declining by about 2%. This is yet another signal that inflation is expected to rise.

Unemployment Likely Higher Than Reported

The May employment numbers were no doubt good news for the U.S. economy. However, we must take a closer look at how it's measured to see if it accurately accounts for employed persons.

Importantly, unemployment is measured using a national survey, not unemployment insurance data. On these surveys, respondents can say they are "unemployed on temporary layoff" or "employed but absent from work for other reasons". This is problematic since many workers who are temporarily laid off would consider themselves "employed but absent from work for 'other reasons'" when in reality they are "unemployed".

Usually, a very low portion of respondents are in this category, but there has been a surge lately (currently about 10X above normal). In fact, if these people are counted for in the unemployment number, it would actually be around 16.3%. The BLS describes this misclassification error in-depth in its recent COVID FAQ report.

There are other issues surrounding the BLS data. The bureau's centers are closed, so interviews are working remotely and are collecting data by phone. This has resulted in a decline in collection rates as well as no in-person interview (due to worker safety). The household survey response rate was 67%, 15% lower than in the period before the pandemic (and 6% lower than in April). Importantly, young people, who have been hit the hardest by unemployment, are notorious for not answering cold phone calls for surveys. This adds further weight to the possibility of an underreported figure.

There are currently 21.5M workers in continuing jobless claims and a total of 42.6M initial claims since early March. There are 157M in today's labor force, so the bare-minimum unemployment rate (using those who have been approved and are receiving funds) is 13.7%. Considering there is a significant backlog and many workers who applied have yet to be approved, we can safely assume unemployment is above 13.7%, but still below the 27% figure using only total initial claims.

It is also worth pointing out that those reporting the more important non-temporary layoffs rose by 15%.

Impact of Ending Subsidies Further Skews Data

There is also the matter of PPP loans. In May, many businesses that laid-off workers realized they would not be able to receive these loans. In May, businesses needed to spend 75% of their aid on wages in order to have loans forgiven so they were encouraged to re-hire workers even if it would cause the business to lose money.

The total value of these loans which have been funded today is about $510B, so we can assume around $400-$450B is used for wages. National compensation before the crash was $9.5T, or $2.38T per quarter (PPP has been in effect for about three months). This tells us that around 20% of the compensation would not be paid if it were not for PPP loans. It is also likely that more loans will be processed due to the program's expansion this month.

Obviously, this is an area of significant uncertainty when trying to find an adjusted unemployment figure. This is partly because we do not know the salary level of those getting an income due to PPP loans. If it is low (likely in the case due to restaurant workers depending on PPP), then the "PPP adjusted unemployment rate" would likely be over 20% higher than it is today. If it is going to high wage earners, then it could be less. An earlier Financial Times report implied unemployment should be closer to 38% without PPP loans, which is in-line with my calculations.

When Good News is Bad News

I believe that it is clear that the unemployment situation today is worse than reported by the BLS. That said, the impact of PPP loans has likely caused true unemployment to be below the 20% expected rate.

I would estimate it is closer to 17% based on the "employed but absent" and decline in survey response errors. The labor participation rate has also declined since March by roughly 5M which are no longer employed people who are not looking for a job. Adding these people to the "official" 13.3%, we get a "U-6" unemployment rate of 21.2%. This still does not account for the estimated 3% undercounting due to misreported persons.

Clearly, the labor market remains in a very weak position. However, that is not the view understood by the market today (as seen by the rally) nor is it the view understood by politicians. Following the report, Trump stated that economist predictions were "the greatest miscalculation in the history of business shows." Though, it seems the miscalculation may at least partially lie with the BLS's system.

Still, this adds further weight to the growing belief that the economy is making a rapid recovery. The left-leaning house recently passed a $3T stimulus bill, "Heroes Act," with another round of $1,200 stimulus checks, but it was unclear if the Senate would approve the bill. However, it was still generally believed that the right-leaning Senate would pass another stimulus bill.

Problematically, the seemingly positive jobs data makes the need for further stimulus seem lower. It lowers the probability of increased stimulus and increases the probability of the $600 weekly benefit will not be extended past July (which has caused most unemployed people's incomes to rise). It has also increased the probability of a Federal Reserve interest rate hike from 0% to 7-8%.

In this sense, good economic news may actually be bearish in the long-run. The economy looks strong almost entirely due to extreme government stimulus. Indeed, evidence suggests stimulus is a direct major reason for the rise in the stock market. If strong economic data arrives, it will lower future stimulus expectations and cause programs to be potentially ended or wound down prematurely.

If the Economy is Rebounding, Inflation Will Skyrocket

Personally, due to the many reasons mentioned above, I do not believe the economy is rebounding yet. However, I believe it will eventually, and when it does the massive increase to the money supply, household incomes, bank cash levels, and savings rates will result in a surge in inflation. Production remains far below pre-crash levels, potentially over 50% below, which means fewer goods will be available even if consumers again feel safe making large purchases. Whenever more money chases fewer items, prices rise.

Indeed, the recent "positive" economic news has resulted in an accelerating increase to the forward inflation rate (calculated using bond prices) as well as a decline in the U.S dollar index. See below:

Data by YCharts

If the U.S dollar declines, import costs rise and result in higher inflation. Of course, higher expected inflation in one county (without a change in short-term rates) also lowers the fair-value exchange rate. This can create a deadly feedback loop that results in large currency declines and potentially hyperinflation.

Indeed, a substantial rise in inflation is actually likely a greater risk to markets than a lasting recession because it can quickly lead to social instability. It will also lead to a crash in fixed-income which most investors increasingly are dependent on today.

My base case is that inflation is on the rise for some goods today, but will not be in full-swing until the economy rebounds. If the economy is indeed rebounding today, inflation could quickly get out of hand. More on this in "Hyperinflation Is Here, We Just Can't See It Yet".

Current Investment Outlook

I remain bearish on long-term bonds as well as richly valued growth/"risk-off" equities such as (in order) technology, utility, and healthcare stocks. In general, the Nasdaq 100 (QQQ) holds a large portion of these companies (ex. utilities). These are assets that did not decline considerably during the March crash, yet are struggling to keep up with smaller companies with low valuations.

Accordingly, I remain bullish on smaller, low "P/E" companies. In particular, those in the commodity production space that will gain from rising inflation. This includes energy (XOP) and miners (PICK).

Still, despite gains in indices, I am happy to have my portfolio slightly skewed toward short equities as it seems clear investors are not appropriately discounting for the difficult economic reality.

Analyst’s Disclosure: I am/we are short TLT. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.