Free download: If wages fell during a recession

You can download my full paper “If Wages Fell During a Recession” with Dan Houser from the Journal of Economic Behavior and Organization (only free until September 24, 2022).

There is a simulated recession in our experiment. We ask what happens if employers cut wages in response. Although nominal wage cuts are rare in the outside world, some of our lab subjects cut the wages of their “employee”. Employees retaliated against nominal wage cuts by shirking, such that the employers probably would have been better off keeping wages rigid.

We also tried the same thing with an inflation shock that allowed the employer to institute a real wage cut without a nominal wage cut. The reaction to that real wage cut was muted compared to the retaliation against the obvious nominal wage cut.

Inflation was implemented after 3 rounds of the same wage to create a reference point.

I blogged about the experiment previously, so I won’t go into more detail here.

The Great Recession happened when I was an undergraduate. As I started my career in research, the issue of employment and recessions seemed like THE problem to work on. The economy of 2022 is so different from the years that inspired this experiment! Below I’ll highlight current events and work from others on this topic.

Inflation used to be something Americans could almost ignore, and now it’s at the highest level I have seen in my lifetime. Suddenly, people are so mad about inflation that politicians named their bill the Inflation Reduction Act just to make it popular.  

The EWED crew has made lots of good posts on inflation. Although job openings and (nominal) wage increases are noticeable right now, Jeremy explored whether inflation has wiped out apparent wage growth.

More recently, the WSJ reports that real wages are down because inflation is so high. “Wage gains haven’t kept pace with inflation. Private-sector wages and salaries declined 3.1% in the second quarter from a year earlier, when accounting for inflation.”

Firms in 2022 did not just sit back and let real wages get eroded exactly proportional to inflation. But it is also not the case that Americans got a raise of 9% to exactly offset inflation. According to our experiment, there would be outrage if workers were experiencing a nominal wage cut in proportion to the real wage cut they are getting right now.

The high inflation combined with a hot job market makes this current economy hard to compare to anything in our recent history. Brian at Price Theory explained that inflation pressure is coming from both supply and demand factors.

Joey has a nice graph on inflation composition.

Did anyone see this coming? Watch Jim Doti of Chapman University predict high inflation based on the money supply in his forecast back in July 2021.

Lastly, our experiment on wage cuts has been cited in these papers:

Intentions rather than money illusion – Why nominal changes induce real effects

Economic stability promotes gift-exchange in the workplace

Wage bargaining in a matching market: Experimental evidence

Can reference points explain wage rigidity? Experimental evidence

Shocking gift exchange

Job Lock is Still Here

Most Americans are covered by employer-sponsored health insurance, either through their own job or a family member’s. This can make it difficult to switch jobs- the new job might not offer insurance, or might have a worse insurance plan or network- locking people into their current job.

Economists have documented since at least the 1980’s how our insurance system seems to reduce job mobility. Several reforms have tried to improve the situation- COBRA, HIPAA, and most recently the Affordable Care Act.

In a paper published this week, Gregory Colman, Dhaval Dave and I evaluate how the extent of “job lock” has changed over time. In short, we find that job lock remains substantial and the Affordable Care Act doesn’t appear to have done anything to improve the situation. The paper has many tables of regression results, but the pictures tell the basic story:

Trends in job mobility for those with and without employer-sponsored insurance (ESI) using Current Population Survey data

The details differ a bit depending on which dataset and identification strategy we use, but a few things are clear:

  1. Macroeconomic factors are dominant in the short run; mobility falls during recessions like 2001 and 2007, then recovers.
  2. The long run trend has been toward lower job mobility for those with AND without employer-based insurance
  3. Those without employer-based insurance are still much more likely to switch jobs (we find 25-45% more likely)
  4. To the extent that this gap has closed since the year 2000, it has come through falling job mobility for those without employer-based insurance more than rising job mobility for those with employer-based insurance

Why does the Affordable Care Act appear not to have improved things? This remains unanswered, but we conclude the paper with some hypotheses:

In fact, our point estimates suggest that job lock actually got stronger following the ACA. One possible explanation for our finding is that the ACA’s individual mandate made insurance even more desirable by fining the uninsured. Another possibility is that workers continue to value employer-provided health insurance more over time as premiums continue to rise

Notes on Austin and Health Economics

I was in Austin Texas for the first time this week for the first in-person meeting of the American Society of Health Economists since 2019. Some quick impressions on Austin:

  • Austin reminds me of many Southern cities, but Nashville most of all. Both historic state capitals that are booming, lots of people moving in and new infrastructure actually being built, forests of cranes putting up new glass towers. Both filled with bars, restaurants, and especially live music. But even with so much happening and so much being built, they don’t *feel* dense, you can always see lots of sky even downtown.
  • Austin seems to be a bizarre “pharmacy desert”, I think I walked 14 miles all through town before I saw one. Contrast to NYC with a Duane Reade on every block. In fact downtown seemed to have almost no chains of any kind, restaurants included; I wonder if this is just about consumer preferences or there’s some sort of anti-chain law.
  • Good brisket and tacos, as expected
  • Most US cities have redeveloped their waterfronts the last few decades to make them pleasant places to be, but Austin has done particularly well here, many miles of riverfront trails right downtown.

Thoughts from the conference:

Continue reading

Nudging Students to Choose a Major

In one sense, it seems like advice does not work. Advice is often ignored and sometimes even resented. People are going to just do what they want.

And yet, many people were in fact influenced by advice at some point in some situation. Many people can tell you about a mentor they spoke with or a book they read. Somehow, we do indeed need to learn about our environments and make choices about career and health and relationships. So, advice does work, sometimes.

A trivial example is why I stopped putting sugar in my coffee. A random anonymous message board post said that you should stop putting sugar in your coffee and your taste will adjust. “You won’t even miss it,” the anonymous poster told me. From that day forward, I stopped putting sugar in my coffee. I’m healthier and I don’t miss it. I was “nudged”. I was also predisposed to make this healthy decision, and I had sought out advice.

We might overestimate the effectiveness of advice because when people bother to talk about it, they mention the one time it affected them. First, they fail to mention the thousands of messages that had no effect (personally I still eat all kinds of junk food that contain sugar despite getting warnings to stop). And secondly, some decisions (perhaps including my coffee-sugar example) would have been made eventually without the advice event. Even recognizing those limitations, I still believe that messaging works sometimes.

It is tempting to think that, at almost zero cost, you could nudge people into making different decisions, just by sending them messages. There is a growing literature on this topic. Economists like myself are collecting data on whether it works.

One of these papers was just published:

Halim, Daniel, Elizabeth T. Powers, and Rebecca Thornton. 2022. “Gender Differences in Economics Course-Taking and Majoring: Findings from an RCT.” AEA Papers and Proceedings, 112: 597-602.

We implemented an RCT among undergraduate students enrolled in large introductory economics courses at the University of Illinois at Urbana Champaign. Two treatment arms provided encouragement to major in economics. A “prosocial” treatment provided information emphasizing the wide variety of career options and personal benefits associated with the major, while an “earnings” treatment provided information on financial returns. We evaluate the effects of the two treatments on subsequent choices to take another economics course and declaration of the economics major by the end of the student’s junior year using student-level matched administrative data. … Our primary aim is to evaluate whether women can be “nudged” into a major with low-cost, theoretically grounded, encouragement/information interventions.

Our primary sample consists of 1,976 students who were freshmen or sophomores during the focal course.

We find that the average male student receiving either treatment is more likely to take at least one more economics course after the focal course, but there is little evidence of increased majoring. The average woman appears unresponsive to either treatment.

Treated women with better than-expected focal-course performance are nudged to take an additional economics course. The likelihood that a woman takes another course in response to treatment increases by 5.6-5.9%-points with a favorable one-third- grade “surprise”. The hypothesis of treatment effects on women’s majoring, mediated or not, is rejected. Men’s susceptibility to treatment is invariant with respect to focal course performance.

Women did not demonstrate a bias towards a pro-social framing, and men did not demonstrate a bias towards a pro-earnings framing.

The pile of null results for messaging, when it is randomly assigned, is growing. It’s good to see null results get published though.

One of my current projects is related, but with a focus on computer programming instead of majoring in economics.

Is this the peak of inflation?

I think so, though the path back to 2% is a long one. Two months ago I wrote that “the Fed is still under-reacting to inflation“. We’ve had an eventful two months since; last Friday the BLS announced CPI prices rose 1% just in May, and that:

The all items index increased 8.6 percent for the 12 months ending May, the largest 12-month increase since the period ending December 1981

Then this Wednesday the Fed announced they were raising interest rates by 0.75%, the biggest increase since 1994, despite having said after their last meeting that they weren’t considering increases above 0.5%. I don’t like their communications strategy, but I do like their actions this month. This change in the Fed’s stance is one reason I think we’re at or near the peak.

Its not just what the Fed did this week, its the change in their plans going forward. As of April, the Fed said the Fed Funds rate would be 1.75% in December, and markets thought it would be 2.5%. But now the Fed and markets both project 3.5% rates in December.

The other reason I’m optimistic is that the days of rapid money supply growth continue to get further behind us. From March to May 2020, the M2 and M3 supply exploded, growing at the fastest pace in at least 40 years:

Rapid inflation began about 12 months later. But the rate of money supply growth peaked in February 2021, then began a rapid decline. Based on the latest data from April 2022, money supply growth is down to 8%, a bit high but finally back to a normal range. Money supply changes famously influence prices with “long and variable lags”, so its hard to call the top precisely. But the fact that we’re now 15 months past the peak of money supply growth (and have stable monetary velocity) is encouraging. Old-fashioned money supply is the same indicator that led Lars Christiansen to predict this high inflation in April 2021 after successfully predicting low inflation post-2009 (many people got one of those calls right, but very few got both).

Stocks also entered an official bear market this week (down 20% from highs), which is both a sign of excess money no longer pumping up markets, and a cause of lower demand going forward.

Markets seem to agree with my update: 5-year breakevens have fallen from a high of 3.6% back in March down to 2.9% today, implying 2.9% average inflation over the next 5 years. Much improved, though as I said at the top the path to 2% will be a long one- think years, not months. Even the Fed expects inflation to be over 5% at the end of this year, and for it to fall only to 2.6% next year.

What am I still worried about? The Producer Price Index is still growing at 20%. The Fed is raising rates quickly now but their balance sheet is still over twice its pre-Covid level and is shrinking very slowly. The Russia-Ukraine war drags on, keeping oil and gas prices high, and we likely still have yet to see its full impact on food prices. Making good predictions is hard.

While I’m sticking my neck out, I’ll make one more prediction, though this one is easier- Dems are in for a bad time in November. A new president’s party generally does badly at his first midterm, as in 2018 and 2010. But this time the economy will be a huge drag on top of that. November is late enough that the real economy will be notably slowed by the Fed’s inflation-fighting effects, but not so late that inflation will be under control (I expect it to be lower than today but still above 5%). Markets currently predict a 75% chance that Republicans take the House and Senate in November, and if anything that seems low to me.

No in-group bias from financial choices in latest experiment

“How Dictators Use Information about Recipients” is my new project with Laura Razzolini. A working paper is up at SSRN. We use the Dictator Game to measure if people are generous toward others who made a similar choice.

In the first stage of the experiment, every player gets to make their own choice about whether or not to invest in a risky option (called Option B). Players can pick Option A if they do not want to invest.

In the second stage, participants get to decide if they will send any money to another anonymous player. If a “dictator” (the person who determines the final allocation of money) decided to take the risk on Option B in stage 1, would they be more generous toward a counterpart if they know that person also picked Option B?

We explain in our paper why the literature indicates such a form of favoritism could be expected.

Social identity theory is the psychological basis for intergroup discrimination. Economic experiments have created feelings of group identity in various ways, leading to significant effects on behavior. Chen & Li (2009) demonstrate that group identity formation can affect social preferences.

Chen and Li (2009) started by having subjects review paintings by two different modern artists. The subjects were divided into two groups, based on their reported painting preferences. Subjects were informed about their group membership by the experimenter.

The Chen and Li paper has been cited almost 2000 times. Group identity is a topic of interest. Several experimental papers demonstrate that strangers can have team feelings induced quickly with the right procedures. Those team loyalties affect behavior in incentivized tasks.

Group feelings artificially induced in the lab by Eckel & Grossman (2005) influence levels of cooperation and contributions to public goods. Pan & Houser (2013)  induce group identities by asking subjects to complete tasks in groups.  Pan & Houser (2019) found that investors trust in-group members more. The in-group has been induced in several different ways in lab experiments. In this paper, we investigate whether in-group effects arise from making a common financial decision in the first stage of the experiment.

Do you think our manipulation in the beginning affected giving?

Nope. There was no effect. Dictators who chose Option B did not give more to recipients who also chose Option B.

Not every result in the paper is a null result. One piece of information caused a large increase in giving. If we inform the dictator that their counterpart started with less money in the first stage (due to bad luck) then the dictator would give more. Sympathy was inspired, as we predicted, by knowing if a recipient was “poor” in the experiment. Conversely, if dictators are informed that their counterpart is “rich” then they excused themselves from having to give up money to help.

Information about financial choices, at least in our sterile simple environment, neither polarized nor united the participants. The giving with only choice information was higher than giving to “rich” but lower than giving to “poor”. Lastly, we provided all of the information at once. With full information, dictators were still heavily influenced by the starting endowments and choices information had no effect.

Understanding polarization is important. Humans exhibit tribal instincts to not help those who are perceived as different. In our experiment we seem to have found one difference that that people are willing to tolerate or overlook.

See also my Works in Progress blog about polarization and a different experiment.  


Chen, Yan, and Sherry Xin Li. “Group Identity and Social Preferences.” American Economic Review 99, no. 1 (March 2009): 431–57.

Eckel, Catherine C., and Philip J. Grossman. “Managing Diversity by Creating Team Identity.” Journal of Economic Behavior & Organization 58, no. 3 (2005): 371–92.

Pan, Xiaofei, and Daniel Houser. “Why Trust Out-Groups? The Role of Punishment under Uncertainty.” Journal of Economic Behavior & Organization 158 (2019): 236–54.

Pan, Xiaofei Sophia, and Daniel Houser. “Cooperation during Cultural Group Formation Promotes Trust towards Members of Out-Groups.” Proceedings of the Royal Society B: Biological Sciences 280, no. 1762 (July 7, 2013): 20130606.

Unfashionable Investing

Investors such as mutual funds, index funds, and hedge funds tend to pick a particular strategy or asset type and stick with it. It’s what they know, it’s what they’re known for, and making major changes would often create legal difficulties; something marketed as a bond fund can’t suddenly switch to stocks even if they think stocks would do much better. Other types of investors like pension funds, endowments and individuals have more flexibility to change their strategies. These investors tend to chase performance, allocating to types of investments that have performed well recently. This can create fashions, types of investment strategies that become more popular for a few years.

These strategies might involve focus on a certain asset class (stocks / bonds / commodities / private equity / real estate / et c), a certain sector or region within an asset class, a certain factor (value, growth, momentum), et c. It seems like institutional incentives, trend chasing, and FOMO lead people and institutions to over-allocate to strategies that have been successful the last 1-5 years and under-allocate to those that haven’t. Everyone sees something has recently been successful, so they pile into it, which drives up prices and makes it look even more successful for a while; but eventually this drives things to be so clearly over-valued that there’s a crash, and the crash scares people away for years until it becomes clearly undervalued. Most recently 2020-2021 saw people pile into growth/tech stocks and alternatives like SPACs/crypto, but the beginning of Fed rate hikes was the signal that the party is over and people (over?)react by pulling out.

Given this, the ideal strategy is to show up right before the party starts, then leave right at the peak; but no one can time it that well. The possibly realistic alternative is to show up early when no one’s there, then leave right when the party’s getting good (Punchbowl Capital?). Timing and identifying which strategies are too hot and which cold enough (Glacier Capital? Cryo Capital?) is the biggest practical question in how to pull this off. The simplest/dumbest way to do it is to avoid timing decisions entirely and just invest fixed proportions into all strategies; when they’re over-valued your fixed investment doesn’t buy many shares, when they’re under-valued it buys lots. This actually sounds like a decent way to go, but its more buying into the Efficient Market Hypothesis than beating it, can we do better? Here are the types of meta-strategies I’m planning to look into:

  • How variable is the timing of strategy boom/busts? Could you possibly just use fixed numbers of months/years- if a strategy’s been hot this long get out, if its been cold this long get in?
  • Use market share numbers, get in when something gets below a certain % of the market and out when it gets above
  • Use valuation numbers like P/E ratios (seems to work well for the overall stock market, may be harder to measure for some strategies/classes)
  • Flow of funds- is there a rate of change that works as a trigger?
  • Proportion of major institutions allocating to each strategy
  • What looks promising right now along these lines (May 2022)? Without looking at the numbers, the perennial strategies that have been out-of-favor a few years seem like value, emerging markets, and commodities (though commodities might be too hot again just now). These (along with real estate; right now homes seem expensive but homebuilders are cheap and I think commercial is too) all did well after the 2000 tech crash

I’m obviously not the first person to think along these lines; the concepts of the commodity cycle and Shiller’s CAPE are related, and Global Macro and Multistrategy funds do some of this. In the latest AER: Insights, Xiao Yan and Zhang echo Robert Shiller and Paul Samuelson that predicting big things like this is actually easier than predicting little things like the valuation of a specific stock:

Samuelson’s Dictum refers to the conjecture that there is more informational inefficiency at the aggregate stock market level than at the individual stock level. Our paper recasts it in a global setup: there should be more informational inefficiency at the global level than at the country level. We find that sovereign CDS spreads can predict future stock market index returns, GDP, and PMI of their underlying countries. Consistent with the global version of Samuelson’s Dictum, the predictive power for both stock returns and macro variables is almost entirely from the global, rather than country-specific, information from the sovereign CDS market

Ungated version here

But I haven’t actually heard of any fund focused on “unfashionable investing” that considers all asset classes and strategies like this. What institution out there would be capable of saying in 2021 “growth stocks are at bubbly levels, we’re switching to commodities”, or saying in 2022 “commodities are high and growth stocks crashed, we’re switching back”? Please let me know if such an institution does exist, or what else to read along these lines.

Get rich or get famous? Edward Thorp vs Myron Scholes

When finance professors publish papers claiming to find inefficiencies in asset markets, my initial reaction is skepticism. The odds are stacked against them to start since asset markets are mostly efficient. Then even if the inefficiency they found is real, shouldn’t they keep that fact to themselves and get rich trading on it?

But listening to a recent interview with Edward Thorp, I realized I shouldn’t entirely discount the possibility that someone would publish a real inefficiency, even a tradeable one. After all, Myron Scholes and Fischer Black did just that when they published the Black-Scholes model in the Journal of Political Economy. This made them famous on Wall Street and in econ/finance academia, and won Scholes the 1997 Nobel Memorial Prize in Economics.

Thorp explained that he had come up with a similar model years earlier, but instead of publishing it, he started a hedge fund and got rich. He says it makes sense that he didn’t share the Nobel Prize, partly because the Black-Scholes model was better than his, but mostly because you should need to publish and share your ideas with the world to get scientific credit for them; his prize was 20% annual returns at his hedge fund.

Why do some opt to get rich, and others to get famous? I’d say academics’ first instinct is to publish everything rather than put it into practice. But Thorp was also an academic, a math professor. Thorp was already famous for publishing a book about how to beat the house at blackjack by counting cards (which is what I knew him for before this interview), so perhaps he valued additional fame less. But he was also already rich from winning at blackjack and from book sales.

Putting ideas into practice can also bring up unanticipated difficulties. When Myron Scholes finally did start working at a hedge fund in 1994 he saw initial success, but by 1998 it had become an embarrassing blunder that inspired the book “When Genius Failed: The Rise and Fall of Long-Term Capital Management”. Scholes may have been better off sticking to academic fame.

Black-Scholes formula for options pricing. The Efficient Markets Hypothesis says that markets instantly incorporate all public information, but original research like this isn’t public until you publish it, and even then it can take years for market participants to fully incorporate it

Why Many Substance Use Treatment Facilities Don’t Take Insurance

According to the latest data, about one in four facilities doesn’t accept private insurance or Medicaid, and more than half don’t accept Medicare. This makes substance use treatment something of an outlier, since 91% of all US health spending is paid for through insurance. Still, there are many reasons to prefer being paid in cash: insurance might reimburse at low rates, impose administrative hassles, and generally try to tell you how to run things.

Providers generally put up with the hassles of insurance because they see the alternative as not getting paid. But if demand for their services gets high enough that they can stay busy with patients paying cash, they will often try going cash-only. Some try to generate high demand by providing excellent service. Sometimes high demand comes from a growing health crisis, as with opioids.

Demand can also be high relative to supply because supply is restricted. US health care is full of supply restrictions, but in this case I wondered if Certificate of Need laws were playing a role. As we’ve written about previously, CON laws require health care providers in 34 states to get the permission of a government board to certify their “economic necessity” before they can open or expand. But there’s a lot of variation from state to state in what types of services are covered by this requirement; acute hospital beds and long-term care beds are most common. 23 states require substance use treatment facilities to obtain a CON before opening or expanding.

States with Substance Use–Treatment CON Laws in 2020. Created using data from Mitchell, Philpot, and McBirney

How do these laws affect substance use treatment? We didn’t really know- only one academic article had studied substance use CON, finding it led to fewer facilities in CON states. But I’ve studied other types of CON, so I joined forces with Cornell substance use researcher Thanh Lu and my student Patrick Vogt to investigate. The resulting article, “Certificate-of-need laws and substance use treatment“, was just published at Substance Abuse Treatment, Prevention, and Policy. Here’s the quick summary:

We find that CON laws have no statistically significant effect on the number of facilities, beds, or clients and no significant effect on the acceptance of Medicare. However, they reduce the acceptance of private insurance by a statistically significant 6.0%.

Overall I was surprised that CON didn’t significantly affect most of the outcomes we looked at, and appears to be far from the main reason that treatment facilities don’t take insurance. Still, repealing substance use CON would be a simple way to improve access to substance use treatment, particularly since CON doesn’t appear to bring much in the way of offsetting benefits.

Going forward I aim to investigate how these laws affect health outcomes like overdose rates, and to dig more into the text of state laws and regulations to determine exactly what is covered by substance use CON in different states. As the article explains, we identified several errors in the official data sources we were using. This makes me worry there are more errors we didn’t catch, and there are certainly things the sources just don’t specify, like in which states the laws apply to outpatient facilities. So I hope we (or someone else) will have even better work to share in the future, but for now this article is as good as it gets, and we share our data here.

College Major, Marriage, and Children

The American Community Survey began in 2000, and started asking about college majors in 2009, surveying over 3 million Americans per year. This has allowed all sorts of excellent research on how majors affect things like career prospects and income, like this chart from my PhD advisor Doug Webber:

See here for the interactive version of this image

But the ACS asks about all sorts of other outcomes, many of which have yet to be connected to college major. As far as I can tell this was true of marriage and children, though I haven’t searched exhaustively. I say “was true” because a student in my Economics Senior Capstone class at Providence College, Hannah Farrell, has now looked into it.

The overall answer is that those who finished college are much more likely to be married, and somewhat more likely to have children, than those with no college degree. But what if we regress the 39 broad major categories from the ACS (along with controls for age, sex, family income, and unemployment status) on marriage and children? Here’s what Hannah found:

Every major except “military technologies” is significantly more likely than non-college-grads to be married. The smallest effects are from pre-law, ethnic studies, and library science, which are about 7pp more likely to be married than non-grads. The largest effects are from agriculture, theology, and nuclear technology majors, each about 18pp more likely to be married.

For children the story is more mixed; library science majors have 0.18 fewer children on average than non-college-graduates, while many majors have no significant effect (communications, education, math, fine arts). Most majors have more significantly more children than non-college graduates, with the biggest effect coming from Theology and Construction (0.3 more children than non-grads).

In this categorization the ACS lumps lots of majors together, so that economics is classified as “Social Sciences”. When using the more detailed variable that separates it out, Hannah finds that economics majors are 9pp more likely than non-grads to be married, but don’t have significantly more children.

I love teaching the Capstone because I get to learn from the original empirical research the students do. In a typical class one or two students write a paper good enough that it could be published in an academic journal with a bit of polishing, and this was one of them. But its also amazing how many insights remain undiscovered even in heavily-used public datasets like the ACS. We’ve also just started to get good data on specific colleges, see this post on which schools’ graduates are the most and least likely to be married.