empirical research – Page 3

Database of Federal Regulations

Omar Al-Ubaydli and Patrick McLaughlin (both at George Mason University) have an article in the most recent issue of Regulation & Governance documenting their RegData database, which “measures [federal] regulation for industries at the two, three, and four-digit levels of the North American Industry Classification System.” While any attempt to quantify regulations is fraught with problems, as the authors note in their paper, their text-based approach would seem as good a method as any (and superior to some) for providing a numerical measure of regulation that could be used for empirical research. And what’s even better, the data are freely available here. The abstract of the paper reads:

We introduce RegData, formerly known as the Industry-specific Regulatory Constraint Database. RegData annually quantifies federal regulations by industry and regulatory agency for all federal regulations from 1997–2012. The quantification of regulations at the industry level for all industries is without precedent. RegData measures regulation for industries at the two, three, and four-digit levels of the North American Industry Classification System. We created this database using text analysis to count binding constraints in the wording of regulations, as codified in the Code of Federal Regulations, and to measure the applicability of regulatory text to different industries. We validate our measures of regulation by examining known episodes of regulatory growth and deregulation, as well as by comparing our measures to an existing, cross-sectional measure of regulation. Researchers can use this database to study the determinants of industry regulations and to study regulations’ effects on a massive array of dependent variables, both across industries and time.

Now, if only there was such a database of State-level regulations.

Craft Beer in the US: History, Stats and Geography

Ken Elzinga (Virginia) and Carol and Victor Tremblay (Oregon State) have a paper in the latest Journal of Wine Economics titled “Craft Beer in the United States: History, Statistics and Geography.” The paper provides a great overview of the history of the craft brew industry as well as some interesting analysis on the geographic development of the industry. The history section seems to draw heavily on Tom Aticelli’s 2013 book The Audacity of Hops: The History of America’s Craft Beer Revolution, but provides a much more concise summary. And paired with the statistical overview of the beer industry in general and the empirical analysis of the craft brew industry that follows, this paper offers a nice, short primer for anyone interested in the history (and economics) of the craft brew industry in the US. The paper’s abstract follows:

We provide a mini-history of the craft beer segment of the U.S. brewing industry with particular emphasis on producer-entrepreneurs but also other pioneers involved in the promotion and marketing of craft beer who made contributions to brewing it. In contrast to the more commodity-like lager beer produced by the macrobrewers in the United States, the output of the craft segment more closely resembles the product differentiation and fragmentation in the wine industry. We develop a database that tracks the rise of craft brewing using various statistical measures of output, number of producers, concentration within the segment, and compares output with that of the macro and import segment of the industry. Integrating our database into Geographic Information Systems software enables us to map the spread of the craft beer segment from its taproot in San Francisco across the United States. Finally, we use regression analysis to explore variables influencing the entrants and craft beer production at the state level from 1980 to 2012. We use Tobit estimation for production and negative binomial estimation for the number of brewers. We also analyze whether strategic effects (e.g., locating near competing beer producers) explain the location choices of craft beer producers.

Do Medical Marijuana Laws Increase Hard-Drug Use?

According to a recent study by Yu-Wei Luke Chu in the Journal of Law & Economics, the answer is not just “No,” but that medical marijuana laws may actually decrease heroin use as consumers substitute the legal marijuana for heroin. Below is the abstract:

Medical marijuana laws generate significant debate regarding drug policy. For instance, if marijuana is a complement to hard drugs, then these laws would increase the usage not only of marijuana but also of hard drugs. In this paper I study empirically the effects of medical marijuana laws by analyzing data on drug arrests and treatment admissions. I find that medical marijuana laws increase these proxies for marijuana consumption by around 10–15 percent. However, there is no evidence that cocaine and heroin usage increases. From the arrest data, the estimates indicate a 0–15 percent decrease in possession arrests for cocaine and heroin combined. From the treatment data, the estimates show a 20 percent decrease in admissions for heroin-related treatment, although there is no significant effect for cocaine-related treatment. These results suggest that marijuana may be a substitute for heroin, but it is not strongly correlated with cocaine.

Fun (Facts & Fiction) With Numbers: Health Care Edition

The graph below, courtesy of the Kaiser Family Foundation, is featured in a VOX post purporting to explain why your health bills are gettng larger (all in one chart!).

The article focuses on the fact that deductibles have risen so dramatically as a major explanation for why it seems like we’re spending so much more on health care, even as health care expenses have been growing more slowly. There is some truth in the claim, and especially to the argument that people are more careful spending on health care when they have to pay for more of it up front, but there are some serious problems with this chart that can lead one to some pretty wrong conclusions.

First, what the graph doesn’t reflect is that the increase in premiums and the increase in deductibles are not, as the picture would appear, necessarily moving together for the people paying them. These are averages, and averages hide lots of information. Moreover, the graph makes it look like the two are increasing are independent of one another; i.e., that people are paying both 24% more in premiums and 67% more in deductibles since 2010. But that’s not the case. Since the ACA, many employers have moved to high-deductible plans that have lower premiums than the low-deductible plans that were popular pre-2010 (see below). What the graph hides is that people with low-deductible plans have seen higher than 24% increases in premiums while people with high-deductible plans have seen much lower increases in premiums–if not actual reductions in their premiums. What has changed is not necessarily how much people are paying for healthcare, but how they are paying it: in premiums or in deductibles. The graph above fails to show that.

Second, looking more closely at the news release on the Kaiser website, the 67% increase in deductibles is an increase in total deductibles paid–not the increase in the average deductible per employee. It reflects not only any increase in deductibles, but the increase in the number of people who have (higher) deductibles. That’s a pretty sneaky way to inflate the numbers on the graph to make it look like the average person is actually paying that much more. Consider the following two graphs, also from the Kaiser Family Foundation 2015 survey. The table on the left shows that premiums for high deductible plans (HDHP/SOs) are significantly lower than premiums for other types of policies. The table on the right shows that the market share of HDHP/SO plans has increased tremendously since gaining ground in 2006. In fact, to relate this to the first graph above, participation in HDHP/SO plans almost doubled from 2010 to 2015, meaning that 50 points of the 67% increase in deductibles could be attributable solely to more people choosing high deductible plans, specifically because the premiums are so much lower. And what the Kaiser report doesn’t say is how much employers contribute to the HSA plans that often accompany HDHP/SO plans. For some individuals, switching to the HDHP/SO plan may actually reduce their total out-of-pocket expense for health care. So while the original graph makes it look like everyone is paying more, that is likely not true for many people–and certainly not at the rate the original graph might suggest.

Finally, because the first graph is in percentages, it hides even more information that changes the story. Suppose deductibles had been $500 and increased to $1,000 or even $2,000. That’s would be a 100-300% increase! 300%! But that’s only $1,500. Not that $1,500 is chump change, but compared to the average annual premium of $6,251 (see the left-hand table above), that’s just 24%–ironically, about the total increase in premiums over the past five years. Even if that $500 deductible grew at the 67% shown in the first graph (which we know from #2 that it didn’t), the increase in actual out-of-pocket health care costs would have been $335–not quite the cost of two lattes a week.

Mark Twain is famously quoted as saying (and actually quoting Disraeli), “There are three kinds of lies: lies, damned lies and statistics..” I’m not saying VOX (or Kaiser) are lying. But be careful when you see things like VOX’s report about some “fantastic new chart.” It’s far too easy to be misled if you don’t think carefully about the numbers being thrown about.

Bonus: If you’re interested in what the research says about the effects of high deductible plans, RAND has a nice summary site with links to additional resources.

How Federal Student Loans Increase College Costs

A recent paper by researchers at the Federal Reserve Bank of New York shows how increases in federal student loan programs–intended to make college more affordable–actually increase the cost of college. As with other markets, when the supply of money available to pay tuition increases, the price of tuition rises. The abstract reads:

When students fund their education through loans, changes in student borrowing and tuition are interlinked. Higher tuition costs raise loan demand, but loan supply also affects equilibrium tuition costs—for example, by relaxing students’ funding constraints.To resolve this simultaneity problem, we exploit detailed student-level financial data and changes in federal student aid programs to identify the impact of increased student loan funding on tuition. We find that institutions more exposed to changes in the subsidized federal loan program increased their tuition disproportionately around these policy changes, with a sizable pass-through effect on tuition of about 65 percent. We also find that Pell Grant aid and the unsubsidized federal loan program have pass-through effects on tuition, although these are economically and statistically not as strong. The subsidized loan effect on tuition is most pronounced for expensive, private institutions that are somewhat, but not among the most, selective.

But the effects don’t stop with rising tuition. This increased demand for college education also exacerbates income inequality by inflating the supply of college graduates. (See this piece by George Leef for a full overview of both the NY Fed paper and the income inequality effects).

It’s not rocket science. It’s pretty simple supply-and-demand stuff, actually. No matter how good the intentions, policies that ignore these effects tend to do more harm than good. In this case, generous federal student loan programs not only lead to increases in tuition that result in even higher loans, but reduce the earning power of graduates (on average) and decrease their ability to repay those loans. A pretty perverse circle of effects indeed.

10th Annual Conference on Empirical Legal Studies

The 10th Annual Conference on Empirical Legal Studies will be held October 30-31, 2015, at Washington University in St. Louis. Sponsored by the Society for Empirical Legal Studies, and hosted by Washington University School of Law and the Center for Empirical Research in the Law, the conference will bring together hundreds of scholars from law, economics, political science, psychology, policy analysis, and other fields who are interested in the empirical analysis of law and legal institutions. The Call for Papers will be published in April. Submissions are due June 26. See the conference page for more details.

Ex Ante vs Ex Post Licensing

Ralph Siebert has an article in the Journal of Competition Law & Economics on “What Determines Firms’ Choices Between Ex Ante and Ex Post Licensing Agreements,” which looks at the timing of technology licensing agreements around research joint ventures in the semiconductor industry. He finds that expectations about potential patent blocking affect the decision of when to license, as do transaction costs, and technology and product market characteristics. His data don’t include much about the specifics of the licensing agreements, but the results are pretty interesting nonetheless. Below is the abstract:

I investigate whether licensing agreements are an appropriate tool for firms to resolve blocking and hold-up problems in high-tech industries. I use a novel and comprehensive database on licensing agreements as well as detailed firm-level information on revenues and patents in the semiconductor industry from 1989 to 1999. It would be interesting to evaluate the post-1999 time period, but data constraints prevent me from doing so. I estimate a bivariate probit model accounting for endogenous selection. I find that different types of licensing agreements, that is, ex ante and ex post licensing agreements, help firms eventually resolve realized blocking. Firms engage in licensing before inventing a new technology (ex ante licensing) if they believe competitors hold patents that can potentially block the commercialization of their technology. In contrast, firms engage in licensing after inventing the technology (ex post licensing) if other firms hold patents that block the commercialization of the technology. The estimation results also show that firms’ activity in technology and product markets plays an important role in explaining choices between ex ante and ex post licensing agreements. It should be kept in mind that the semiconductor industry is high-paced and the data patterns might have changed after 1999.