The passing of Alan Krueger last week is not the first of a top-level economist that I have had to read, and it certainly will not be the last. But it will most palpably place itself among the more startling deaths of a prominent scholar in the field that I have and will have come to hear; and this not only because he had barely made past the halfway point of his fifty-ninth year on the morning of March 16th, 2019, but also because of what the first report of the cause of his death succinctly stated—death by suicide.
Besides knowing him as an economist at Princeton University my familiarity with the works of Professor Krueger was vague, even forgettable one could add without my retort. It was only a few years after first encountering his name, in a course I took during my masters, that I was re-introduced to him more rigorously, the manner an economist wishes to be befriended. The Course? –Econometric Methods. The paper? –Minimum Wages and Employment, by David Card and Alan B. Krueger, 1994. Over these twenty-five years since the paper first appeared in a top journal, its relevance has only gained a tighter grip of the hands of policy-makers; and has been embraced and replicated by subsequent generations of eager empiricists.
Basic economic theory assertively predicted that raising the minimum wage above the existing wage in a competitive market would lead to rising unemployment. The reasoning can be easily accessed from chapter 1 of any first-year undergraduate economics textbook. ‘Demand and supply of workers’ determine the existing minimum wage rate. Any attempt to artificially increase the wage would lead to layoffs by firms, all else remaining equal. Well, Professor Krueger’s paper (with Card) showed that raising the minimum wage not only didn’t cause unemployment (among the fast-food restaurants they studied) as predicted by theory but raised employment slightly. So what was such a well-established and appreciably-preached theory to say of this disconcerting outcome?. . . . . Well, go back to what it says. People often ignore the rather nuanced yet powerful assumption “all else remaining equal”. Higher minimum wages will lead to higher layoffs conditional on all other factors staying the same. But firm behaviors could change, worker preferences may not remain constant, there could be spillover effects, or labor markets may not be competitive to start with. It was right, then, to scrutinize assertions that sought help from the simple and static “demand and supply model” as being, well, over-simplified. Though the authors did not dive deep into the reasons for the surprising results, they helped usher in bigger debates and approaches to the minimum wage in the US.
Here are the circumstances that neatly aligned with the path of the authors that made for a compelling case towards their findings: a controlled experiment. In 1992, after a contemptuous battle, New Jersey lawmakers raised the minimum wage from $4.25 to $5.05—a 19% increase. Now to simply compare the level of employment in New Jersey after and before the legislative change would be erroneous; because it would verily omit seasonal effects and trends that mark their effect over time. Even without the legislative intervention, there should be little scientific reason to discard beliefs that fast-food employment in New Jersey should have gone up. Card and Krueger navigated their way around this problem through a modeling technique called ‘Difference-in-difference’. Neighboring state Pensylvania did not change their wage laws. So by using a sample of fast-food restaurants from Pennsylvania, Card and Krueger were able to establish more sound results by comparing before-and-after employment rates in both states. To produce robust results, a strong assumption taken here was that both states experienced the same parallel trends over time (or at least during the period of the study). The model’s results showed no inkling towards the belief that raising the minimum wage led to unemployment. One can imagine the intellectual blow the paper must have handed to the obstinate faces of hardliners and ideologues.
Alan Krueger was also a devoted lover of music, and not with a dilute and pretentious attempt at it but one with genuine passion, so deep that it was quite able to seduce his pen toward it, of which when we speak we mean his academic works on the music industry’s operations. In an unusual paper titled “Rockonomics: The economics of popular music”, Alan Krueger, with co-author Marie Connolly, outlined a detailed study of the changes that were taking place in the music industry at the turn of the 20th century. Using data from Pollstar and The Rolling Stone Encyclopedia of Rock & Roll, the authors observed that a vast chunk of the incomes of artists came from concert revenues rather than record sales. This differential ratio stood on average at about 7.1 to 1 among the 35 best selling artists who toured in 2002. This was puzzling because in 2003 the total value of recording sales (including CDs, singles, LPs, etc.) in the U.S. was $11.8 billion, while the total value of concert ticket sales was $2.1 billion. Well, the reason, the authors stated, was that most of the revenues from record sales went to the recording companies and publishers, leaving the artists with an abysmal share. Artists were often tied up to contracts that were designed to extract every ounce of surplus available, even once causing the wife (we shall call her Sharon) of one prominent rockstar to headbutt the band manager in protest.
When looking at price changes, the authors observed that both the price of concert tickets and the CPI (or Consumer Price Index, which measures inflation) grew at roughly the same rate from 1981 up until 1997. Then a divergence appeared. Between 1997 to 2003, concert ticket prices witnessed a 2 fold increase while CPI figures didn’t show any change in trend, growing at roughly 3% per year. These results held whether looking at an artist’s average concert ticket price over time, or a venue’s average ticket price over time. Three trends observed are worth mentioning. First, the number of shows performed showed an inverse V, rising in the 80s, peaking in the 90s, and declining thereafter. Second, the number of concert tickets sold remained flat for most of the 80s and 90s but declined sharply from 2000 onwards. Third, despite flat sales in the 80s and 90s, concert revenues went up until 2000, implying the effect of the significant price increase, but decreased since, further implying that the price increase after 2000 did not weather off the effect of the ticket sales drop.
Krueger and Connolly proposed three hypotheses for these rather odd trends. Music concerts are a low-productivity sector. It takes just as much effort to organize a concert today as it did a generation ago. So to make up for this deficiency, concert prices had to rise faster than other prices. Monopoly is another reason the authors lay out, though more cautiously. In the late 90s, giant multimedia conglomerates increased their share of ownership of radio stations, venues and billboards, in a way granting them the power to force prices up and take home a sizable portion of the revenues. But the most compelling hypothesis involves monopoly power with complementary goods. In most concerts, records are sold in stalls located around the venue. So, tickets and records work as complements. Second, both markets operate, in some sense, under a monopoly structure. A simple model they incorporated predicted that if artists believed that greater attendance would boost record sales, they could accommodate a lower price (below the monopoly price level) for tickets. But with the emergence of file sharing and other technological innovations in the early 2000s, there were faint reasons to assume that records would stand as a strong complement to concerts. This, their model showed, would shoot up concert prices to the monopoly price.
Professor Alan Krueger’s deep desire to get into the root of a problem was a trait everyone admired. He was always hungry for his craft, up to his very last days. A book on rock music he was working on will be posthumously released in the summer. Condolences swiftly rushed to the horrific news of his death, from colleagues he worked alongside with, to former presidents he dutifully served. He left us all too soon, no one will contest that. But “raising the minimum wage is bad”? I know a bunch who will immediately position themselves on the other side of the debate floor. That is a lesson Alan Krueger taught me, that one can exude influence not by being the loudest in the room but by developing a sincere yearning for the truth, and then reckoning the most salient way of disseminating it. Young minds will swallow his prolific works for decades to come, and older minds will pause for him, some with envy, others with awe, but all of whom will harbor alongside undying respect. He was a fan of Bruce Springsteen, and the percussion-charged lyrics of this song by his fellow New Jerseyan quite aptly helps transmute our thoughts into music:
You can’t start a fire
You can’t start a fire without a spark
This gun’s for hire
Even if we’re just dancing in the dark
Alan Krueger lit the spark for a flame that still burns bright, and because of that, we all can see more clearly.
Author: Resem Makan
Resem Makan is an Economics PhD student at the University of Washington. Before this, he was at the Indian Statistical Institute (Delhi) where he studied Quantitative Economics. He grew up in Nagaland, Aizawl, and Shillong, and thus feels a part of everywhere.