Debate Club

Pop quiz. What’s the first rule of Debate Club?

Surprisingly, it’s not “do not TALK about DEBATE CLUB!”.

It turns out the first rule of Debate Club is “read all your preparatory material”. Or at least that’s what I presume it is. I based that presumption on memories of my high-school Debate Club friends who spent their free-study hours furiously highlighting reams of printed journal articles they’d found on microfiche in the school library. Yes, I realize I’m dating myself with the microfiche reference.

I know what you’re thinking, and I agree — the Fight Club reference sounds way cooler. Then again, what do I know? I spent the majority of my time in high school cutting weight for the real fight club (the wrestling team). Which, by the way, no one ever talked about … probably because no one ever attended high school wrestling matches (other than parents). While my choice of extracurricular activities may have prepared me well for the lonely life of a blogger with three devoted readers, it probably didn’t prepare me well for a debate on investing the Emergency Fund (EF) against some of the heavy hitters in the Financial Independence (FI) community.

EF Preparatory School

Speaking of preparation, did you read part one of this series yet? Based on the fact that it didn’t go viral, I’ll presume you didn’t. You should though. Not that it’s a classic or anything, but it will at least prepare you for the upcoming debate. It also establishes what I think is a level playing field for the contest that follows.

While you’re cramming (or cribbing) for the big show, you may also want to listen to ChooseFI Episode 66 and 66R if you have the time. If not, you could probably just skim (or print out and highlight?) Big Ern’s series of articles at his website Early Retirement Now; where he explains why he invests his EF in equities. He essentially repeats his arguments from those articles in the ChooseFI episodes. Given this fact, and unless otherwise noted, I’m going to reference Ern’s articles, rather than the ChooseFI episodes for the remainder of this debate. Once you’re done or feel prepared enough to judge the debate on its merits, go ahead and join me in the next section.

Other Debate Club Rules

In part one of this series, I pointed out that traditionally the EF is defined as:

… an account for funds set aside in case of the event of a personal financial dilemma, such as the loss of a job, a debilitating illness or a major repair to your home. The purpose of the fund is to improve financial security by creating a safety net of funds that can be used to meet emergency expenses as well as reduce the need to draw from high-interest debt options, such as credit cards or unsecured loans.

Although neither the ChooseFI Podcast hosts, Brad and Jonathan, nor Big Ern chose to define the EF in those terms; that is the definition we’ll stick with for this debate. I think that’s fair. After all, if you’re going to admittedly attempt to slay the Emergency Fund sacred cow, then you need to slay the sacred cow as defined by the community who worships it — not as you choose to define it for the convenience of your argument.

I had a stodgy old English professor in my master’s program fail my World War I essay for re-framing his question, so I’m doing the same to the ChooseFI guys. Therefore, any points Brad, Jonathan, and Ern made during their two episodes, or blog articles, that infer an advantage to their argument through their redefined definition of an EF, are disqualified. Sorry guys, but you can’t have your cake and eat it too. If you want to re-argue your points under the agreed upon definition of the EF, then feel free to hit the comment button below, or submit a guest post.

With that said, I’m not simply going to cherry pick Ern, Brad, or Jonathan’s weakest remaining arguments. Instead, I’m going to pick their strongest remaining arguments to run a point/counter-point debate against. I’m going to do that because I’m truly curious to see what, if anything, about the ChooseFI and the Early Retirement Now 100% certified recommendation to invest the EF, actually survives scrutiny. I’m dividing the point-counterpoint debate into rounds, each of which will be its own blog post. If I didn’t run it this way, I’d have a 10,000-word post before I knew it. Nobody wants that. I’ll leave the judgments to the audience as to who wins each round. Let me know in the comments section what you think.

Finally, I just want to remind everyone we are all internet friends here. Let’s keep the vitriol and below the belt hits to a minimum. Now, I want everyone to return to their corners, and when I ring the bell, come out swinging!

Round 1: Probability

The strongest argument made in Ern’s articles, and both of the ChooseFI episodes, is that the probability of needing an EF for its true purpose (job loss) is low. Furthermore, according to Ern, there’s a low probability of correlation between the likelihood of a person needing an EF (from a job loss) at the same time the stock market tanks. As he states in his article “Top 10 reasons for having an emergency fund – debunked (Part 1)“:

True. There is a positive correlation between job losses and both the business cycle and stock returns. But the correlation is quite weak. Only 17% of unemployment claims occur during recession periods, 83% during expansions since the government started measuring those in 1967 (Weekly Unemployment Claims). 17% is higher than the recession probability since 1967 (13%), hence the positive correlation, but you are still five times more likely to claim unemployment benefits during an expansion than during a recession.

By the way, Ern’s link (above) from his original article should take you to where he pulled the raw data to calculate the 17% figure. However, the link does not always work. There were several time this weekend as I wrote the article that the link took me to the website, but splashed the “Ooops” error. I verified that the data exists but didn’t run the calculations for myself. Ern is something of a polymath; I trust him, so I left the link in my quote for accuracy’s sake.

What is Ern’s quote above trying to convey anyway? Well, recessions and Bear markets (downturns of 20% or more for the stock market) tend to occur together. As Gregory M. Drahuschak pointed out as recently as September 2017, “since 1929 there have been 16 bear markets that in all but one period coincided with a recession – a period loosely defined as two or more consecutive quarters of negative GDP growth”. Despite this, Ern assesses the risk to the money from the stock market invested EF is low.

Ern’s “low risk” argument appears based on a few different facts. First, Bull markets run much longer than Bear markets. The historical average runs roughly nine years of Bull market for every 1.4 years of Bear market. Thus, with a relatively weak positive correlation in jobless claims during a recession (and a Bear market), a person is simply much more likely to actually lose their job in a year when the market is up. Therefore an invested EF is, statistically, far more likely to have experienced gains, rather than losses, at the time a person must utilize it.

This fact is backed up by Bureau of Labor Statistics (BLS) analysis of recession and jobless data which shows job loss is a lagging indicator of a recession, not a leading indicator. This means that even if a person loses their job from a recession, they are more likely to lose it at the end of the recession, or after the actual recession (and Bear market) has ended. In fact, according to BLS analysis since 1969, job loss as a result of a recession tends to peak at the end or after the recession ends. That trend is plainly apparent in the below BLS chart.

This, again, means that an EF invested in equities is not only likely to be intact when needed but actually larger than when first invested. In fact, Ern goes so far as to point out that:

The entire emergency fund rationale thus depends on the crazy assumption that the large cash stash sitting around would have been optimal under very specific and unlikely circumstances, namely … losing your job exactly at the bottom of the stock market…  Rational people would weigh the pros and cons of the emergency fund over all possible outcomes. And just to be sure, you can and should value the liquidity of an emergency fund at the bottom of the crisis more than the opportunity cost of the emergency fund during normal times to account for risk aversion. But unless you are crazy, crazy risk-averse, the emergency fund is still not worth it!  (*emphasis Ern’s)

Round 1 Rebuttal: The Nuance of Probabilities

Ern’s right, rational people would weigh the pros and cons over all the scenarios, but he hasn’t. He’s citing averages to make a general point, which I concede is strong at face value. Yet, much like I pointed out in my previous article when I discussed risk tolerance vs. risk capacity, what’s true for one person isn’t for another. Ern hasn’t actually peeled back the onion on the correlation of job loss in specific industries due to a recession. If he had, he would’ve seen the picture gets much more convoluted.

The BLS analyzes recessions and job loss … a lot. Included in their body of work are numerous studies that break down job losses during recessions by all sorts of demographics. Turns out your age, sex, state (or country) of residence, education level, and ethnicity all impact your probability for lay off in various ways. The trends in those demographics are not always consistent from recession to recession. However, there are some areas of consistency … especially among the industries typically hit hardest by a recession. For instance, consider this graphic that I found in a BLS report from 2012:

 

If that’s not enough, you could always go to this article from 2010 at the Huffington Post. The bottom line is that the picture is basically the same wherever you look; some professions experience major job losses due to a recession, some don’t, and strangely, others benefit. Construction and manufacturing are the consistent losers during U.S. recessions stretching back to the 1970s. Although not 100% consistent, the retail sector also tends to see major job loss due to a recession. This data is supported by the BLS’s examination of the number of unemployed persons per job opening from the Great Recession as well.

Unfortunately, the data isn’t precise enough to indicate which industries typically lead job loss, and which others lag, when a recession hits. However, both the manufacturing and housing sectors are leading indicators for economic health. As a result, it’s probably a safe assumption that workers in those industries feel the effect of job loss first during a recession.

There’s another interesting nuance worth considering in respects to this debate, and it’s how long someone will stay unemployed once they lose their job as a result of a recession. Turns out that Nate Silver’s website, FiveThirtyEight.com, did some serious statistical analysis on this subject. The results speak for themselves:

A FiveThirtyEight analysis shows that by far the single biggest predictor of whether someone will be out of work for a year or more is the state of the economy when he or she loses his or her job.¹ Over the past 15 years, a period spanning two recessions, a one-point increase in the unemployment rate increased an individual’s odds of remaining unemployed for at least a year by about 35 percent. No other characteristic — age, sex, race, marital status, education or occupation, among others — had even close to that big an effect.

In English, this means if a person loses their job when the unemployment rate is high, there’s a much higher probability they will remain unemployed for at least a year. Now, before Ern comes back with the retort that the extension of unemployment benefits for longer periods of time (up to 99 weeks in some cases during the Great Recession) might explain the above findings; the FiveThirtyEight.com author actually tested for this, and found the opposite to be true.

The BLS data backs up the above analysis as well. The BLS states in their study, Trends in Long-Term Unemployment, “as the duration of unemployment increases, the likelihood of becoming employed in the following month declines.”. Apologies for the atrocious English in that quote, apparently Economists don’t speak English good. However, they build great charts, like the one below which displays this issue graphically. Notice how the teal colored section grows bigger over time. That’s the visual version of the effect described above.

 

What’s the takeaway here? Despite Ern’s seemingly strong case, the devil, of course, is in the details. If you work in construction, manufacturing, or even retail, the probability of losing your job due to recession is higher than the averages cited by Ern. How high? Unfortunately, I can’t find anything specific, but the data makes it appear orders of magnitude larger than other job sectors.

Furthermore, the construction and manufacturing sectors tend to act as leading indicators for the economy; in both expansion and contraction. This increases the likelihood that workers in those sectors lose their jobs on the front end of the unemployment spike — during a recession — vice after. In other words, during a Bear market when stocks traditionally tank. Thus, the “very specific and unlikely circumstances” that Ern rails against in his article appears a lot more likely for anyone who works in one of those fields. As a result, I’d say in rebuttal to Ern, for people who work in certain industries, the standard advice to keep an EF in some form of cash holding, for an amount that could cover expenses for up to and including a year, may not be so “crazy” after all.