orgtheory.net

Archive for the ‘economics’ Category

let’s hear it for null results

A common, and important, critique of journals is that they don’t want to publish null results. So when I saw a new piece in Socio-Economic Review yesterday reporting essentially null findings, I thought it was worth a shout-out. The article, by economist Stefan Thewissen, is titled, “Is It the Income Distribution or Redistribution That Affects Growth?” (paywalled; email me for a copy). Here’s the abstract:

This study addresses the central question in political economy how the objectives of attaining economic growth and restricting income inequality are related. Thus far few studies explicitly distinguish between effects of income inequality as such and effects of redistributing public interventions to equalize incomes on economic growth. In fact, most studies rely on data that do not make this distinction properly and in which top-coding is applied so that enrichment at the top end of the distribution is not adequately captured. This study aims to contribute using a pooled time-series cross-section design covering 29 countries, using OECD, LIS, and World Top Income data. No robust association between inequality and growth or redistribution and growth is found. Yet there are signs for a positive association between top incomes and growth, although the coefficient is small and a causal interpretation does not seem to be warranted.

Okay, so there’s the “signs for a positive association” caveat. But “the coefficient is small and a causal interpretation does not seem to be warranted” seems pretty close to null to me.

In light of the attention this report from S&P has been getting — e.g. from Krugman today (h/t Dan H.) — all solid findings, null and otherwise, on the inequality-growth relationship warrant publication. Hats off to SER for publishing Thewissen’s.

 

Written by epopp

August 8, 2014 at 4:35 pm

grade inflation experiment

A recent article in the Journal of Economic Perspectives reports a recent attempt to curb grade inflation. High GPA departments at Wellesley College were required to cap high grades. The abstract:

Average grades in colleges and universities have risen markedly since the 1960s. Critics express concern that grade inflation erodes incentives for students to learn; gives students, employers, and graduate schools poor information on absolute and relative abilities; and reflects the quid pro quo of grades for better student evaluations of professors. This paper evaluates an anti-grade-inflation policy that capped most course averages at a B+. The cap was biding for high-grading departments (in the humanities and social sciences) and was not binding for low-grading departments (in economics and sciences), facilitating a difference-in-differences analysis. Professors complied with the policy by reducing compression at the top of the grade distribution. It had little effect on receipt of top honors, but affected receipt of magna cum laude. In departments affected by the cap, the policy expanded racial gaps in grades, reduced enrollments and majors, and lowered student ratings of professors.

My sense is that this shows that grade inflation, whatever its historical origins, acts as a competitive advantage for programs that few other market advantages. If you don’t have a strong external job market or external funding, then you can boost enrollments via grade inflation. It also absolves programs by masking racial under performance. The lesson for academic management is this: If you have inequality in funding, departments will compensate by weak grading. If you have inequality by race, departments will compensate by weak grading. Thus, academic leaders who care about either of these issues should implement policies where departments don’t choose standards and are accountable for results.

50+ chapters of grad skool advice goodness: Grad Skool Rulz/F

Written by fabiorojas

August 5, 2014 at 12:01 am

is sociology a poor source of policy stories?

A few years ago, I bought a copy of Charles Tilly’s Why?, just for fun sociology reading. All the Important sociology reading got in the way, and I never read Why?

But while I was unpacking this week I came across it and thought I’d bring it along on a car ride to Providence over the weekend. Not only is it a fun read, as well as touchingly personal at times, it turned out to be surprisingly relevant to stuff I’ve been thinking about lately.

The book is organized around four types of reasons people give for things…any things: their incarceration in mental hospitals, why a plane just flew into the World Trade Center, whether the last-minute change of an elderly heiress’s will should be honored. In grand social science tradition, the reasons are organized into a 2 x 2 table:

Popular Specialized
Formulas Conventions Codes
Cause-Effect Accounts Stories Technical Accounts

 

Why? illustrates these types with a wide range of engaging examples, from eyewitness accounts of September 11th to the dialog between attending physicians and interns during hospital rounds.

Conventions are demonstrated by etiquette books: they are reasons that don’t mean much of anything and aren’t necessarily true, but that follow a convenient social formula: “I lost track of the time.” Stories are reasons that provide an explanation, but one focused on a protagonist—human or otherwise—who acts, and which often contain a moral edge: evangelist Jerry Falwell’s account of how he came to oppose segregation after God spoke to him through the African-American man who shined his shoes every week. Both conventions and stories are homely, everyday kinds of reasons.

Codes and technical accounts, on the other hand, are the reasons experts give. Reasons that conform to codes explain how an action was in accordance with some set of specialized rules. The Department of Public Works did not repair the air conditioning because they lacked a form 27B/6. While law is the quintessential code, Tilly shows that medicine follows codes to a surprising extent as well.

Finally, technical accounts attempt to provide cause-effect explanations of why some outcome occurs. Jared Diamond argues that Europe developed first because it had domesticable plants and animals and sufficient arable land, and lacked Africa’s north-south axis. Technical accounts draw on specialized bodies of knowledge, and attempt to produce truth, not just conformity with rules.


 

I’ve spent a lot of time in recent months thinking about what experts do in policy, and thinking about the different paths through which they can have effects. Lots of these effects are technical, of course. Expert opinion may not determine the outcome in debates over the macroeconomic effects of tax policy changes or what standards nutrition guidelines should be set at, but there’s no question that they’re informed by technical accounts.

But at least as important in influencing a wider audience are the stories experts can tell. Deborah Stone wrote about these “policy stories” decades ago, though she wasn’t especially focused on experts’ role in creating them. Political scientists like Ann Keller, however, have shown that scientists, too, translate their expertise into policy stories—for example, that human activity was creating the sulfur and nitrogen oxides that produce acid rain, destroying fisheries and making water undrinkable. These stories are grounded in technical accounts, but are simplified versions with moral undertones that point toward a particular range of policy solutions—in this case, doing something about the SOx and NOx emissions that the story identifies as creating the problem.

Some kinds of expertise, or rather some kinds of technical accounts, are more amenable than others to translation into policy stories. Economic models, in particular, are often friendly to such translation. For example, although this isn’t the language I use there, my book in part argues that U.S. science policy changed because of a model-turned-story. Robert Solow’s growth model, which includes technology as a factor that affects economic growth (by increasing the productivity of labor), became by the late 1970s the basis of a powerful policy story in which the U.S. needed to improve its capacity for technological innovation so that it could restore its economic position in the world.

Similarly, a basic human capital model in which investment in training results in higher wages easily becomes a story in which we need to improve or extend education so that people’s income increases.

Sociological models, even the formal ones, seem less amenable on average to these kinds of translations. Though Blau and Duncan’s well-known status attainment model could be read as suggesting education as a point of intervention to improve occupational status, it seems fairer to read it as saying that occupational status is largely determined by your father’s occupation and education. While this certainly has policy implications, they are not as natural an extension from the model itself. It hearkens back to that old saw—economics is about how people make choices; sociology is about how they don’t have any choices to make.

Blau & Duncan

I guess part of the appeal of Why? for me was that it mapped surprisingly well onto these questions that were already on my mind. Mostly I’ve thought about this in the context of economic models becoming policy stories. I wonder, though, whether my quick generalization about the technical accounts of sociology lending themselves less readily to compelling policy stories actually holds up. What are the obvious examples I’m missing?

Written by epopp

July 9, 2014 at 7:00 pm

higher education readings: about students

This week, some readings on who goes to college and why:

Use the comments for more suggestions.

50+ chapters of grad skool advice goodness: From Black Power/Grad Skool Rulz

Written by fabiorojas

June 27, 2014 at 12:01 am

why sociology is a bit more threatening than economics: the case of fetal alcohol syndrome

Consider the following approaches to the same in issue – fetal alcohol syndrome (FAS). In 2000, Elizabeth Armstrong and Ernest Abel published an article in the journal Alcohol and Alcoholism arguing that fetal alcohol syndrome had become a moral panic.  Even though people had become obsessed with FAS, there was actually very little evidence to suggest that moderate alcohol consumption damaged fetuses. This argument is elaborated in the 2008 book Conceiving Risk, Bearing Responsibility. In 2013, the economist Emily Oster published a book called Expecting Better, which assesses pregnancy advice with a review of the pertinent clinical evidence. Like Armstrong, Oster finds that the norm against moderate alcohol consumption is  not supported by the data.

The comparison between Oster and Armstrong is revealing. For example, more people know about Expecting Better because, frankly, economists are more respected than sociologists. But there is a deeper lesson. When Oster frames her work, she presents it as a morally neutral project. Her framing is roughly: “Statistics is hard, people may not have all the facts, and you might have a mistaken belief, but as an economist,  I am trained in statistics. I can help you make a better choice.” Thus, the reader is morally blameless.

In contrast, Armstrong’s approach to FAS relies on standard explanations of moral panics in sociology. It goes something like this: “The facts we believe reflect our underlying biases. These biases reflect our evaluations of certain types of people, who may not deserve that stigma.” Thus, if the reader buys FAS, they are implicated in an immoral action – unfairly exercising gender prejudice. Heck, all of society is implicated.

This is an interesting observation about the public image of disciplines. Economists may advocate unpopular policies (e.g., they are often critical of minimum wage laws) but their moral framework is fairly neutral and technocratic. If you don’t buy my policy, it’s probably because you aren’t aware of all the factors involved. You haven’t calculate the social welfare function properly! In contrast, sociologists often make arguments that implicate the moral character of the audience. And that doesn’t buy you a lot of friends.

50+ chapters of grad skool advice goodness: From Black Power/Grad Skool Rulz

Written by fabiorojas

June 11, 2014 at 12:01 am

a rant on ant

Over at Scatterplot, Andy Perrin has a nice post pointing to a recent talk by Rodney Benson on actor-network theory and what Benson calls “the new descriptivism” in political communications. Benson argues that ANT is taking people away from institutional/field-theoretic causal explanation of what’s going on in the world and toward interesting but ultimately meaningless description. He also critiques ANT’s assumption that world is largely unsettled, with temporary stability as the development that must be explained.

At the end of the talk, Benson points to a couple of ways that institutional/field theory and ANT might “play nicely” together. ANT might be useful for analyzing the less-structured spaces between fields. And it helps draw attention toward the role of technologies and the material world in shaping social life. Benson seems less convinced that it makes sense to talk nonhumans as having agency; I like Edwin Sayes’ argument for at least a modest version of this claim.

I toyed with the possibility of reconciling institutionalism and ANT in an article on the creation of the Bayh-Dole Act a few years back. But really, the ontological assumptions of ANT just don’t line up with an institutionalist approach to causality. Institutionalism starts with fairly tidy individual and collective actors — people, organizations, professional groups. Even messy social movements are treated as well-enough-defined to have effects on laws or corporate behavior. The whole point of ANT is to destabilize such analyses.

That said, I think institutionalists can fruitfully borrow from ANT in ways that Latour would not approve of, just as they have used Bourdieu productively without adopting his whole apparatus. In particular, the insights of ANT can get us at least two things:

1)      It not only increases our attention to the role of technologies in shaping organizational and field-level outcomes, but ANT makes us pay attention to variation in the stability of those technologies. It is simply not possible to fully accounting for the mortgage crisis, for example, without understanding what securitization is; how tranching restructured, redistributed and sometimes hid risk; how it was stabilized more or less durably in particular times and places; and so on.

You can’t just treat “securitization” as a unitary explanatory factor. You need to think about the specific configuration of rules, organizational practices, technologies, evaluation cultures and so on that hold “securitization” together more or less stably in a specific time and place. Sure, technologies are sometimes stable enough to treat as unified and causal—for example, a widely used indicator like GDP, or a standardized technology like a new drug. But thinking about this as a question of degree improves explanatory capacity.

An example from my own current work: VSL, the value of a statistical life. Calculations of VSL are critical to cost-benefit analyses that justify regulatory decisions. They inform questions of environmental justice, of choice of medical treatment, of worker safety guidelines. All sorts of political assumptions — for example, that the lives of people in poor countries are worth less than people in rich ones — are baked into them. There is no uniform federal standard for calculating VSL — it varies widely across agencies. ANT sensitizes us not only to the importance of such technologies, but to their semi-stable nature—reasonably persistent within a single agency, but evolving over time and different across agencies.

2)      Second, ANT can help institutionalists deal better with evolving actors and partial institutionalization. For example, I’m interested in how economists became more important to U.S. policymaking over a few decades. The problem is that while you can define “economist” as “person with a PhD in economics,” what it means to be an economist changes over time, and differs across subfields, and is fuzzy around the borders.

I do think it’s meaningful to talk about “economists” becoming more influential, particularly because the production of PhDs happens in a fairly stable set of organizational locations. But you can’t just treat growth theorists of the 1960s and cost-benefit analysts from the 1980s and the people creating the FCC spectrum auctions in the 1990s as a unitary actor; you need ways to handle variety and evolution without losing sight of the larger category. And you need to understand not only how people called “economists” enter government, but also how people with other kinds of training start to reason a little more like economists.

Drawing from ANT helps me think about how economists and their intellectual tools gain a more-or-less durable position in policymaking: by establishing institutional positions for themselves, by circulating a style of reasoning (especially through law and public policy schools), and by establishing policy devices (like VSL). (See also my recent SER piece with Dan Hirschman.) Once these things have been accomplished, then economics is able to have effects on policy (that’s the second half of the book). While the language I use still sounds pretty institutionalist—although I find myself using the term “stabilized” more than I used to—it is definitely informed by ANT’s attention to the work it takes to make social arrangements last. Thus I end up with a very different story from, for example, Fligstein & McAdam’s about how skilled actors impose a new conception of a field — although new conceptions are indeed imposed.

I don’t have a lot of interest in fully adopting ANT as a methodology, and I don’t think the social always needs to be reassembled. The ANT insights also lend themselves better to qualitative, historical explanation than to quantitative hypothesis testing. But all in all, although I remain an institutionalist, I think my work is better for its engagement with ANT.

Written by epopp

June 5, 2014 at 8:31 pm

econobros: that’s the end of the chain!

In one of my graduate courses, I taught the Rand health insurance experiment. It’s a famous study where some people were randomly given health insurance coverage to see how it affected access and health. The bottom line is that using insurance to decrease the costs of health via low co-payment helps with access, but not with health. In the discussion, I mentioned how this result surprises people. Then, one of my BGS* said the following, paraphrased by me:

The reason this might be surprising from an economic perspective is that social behavior is a question of relative prices. Obviously, purchasing health care would become more common if it were made easier. However, health is often beyond the ability of individuals to directly influence. Health might be due to genetic factors, social class, occupation, and other processes that are not easily countered by a visit to a doctor. Health is the result of a long chain of events. These policy interventions only happen at the end, so the modest effects shouldn’t be surprising.

Now, we did discuss the famous finding that the intervention helped with low-income individuals. But this supports the “end of the chain” view of health. For most people, they already have the resources and environment that will help with prevention of chronic health problems (e.g., malnutrition in youth) or managing short term issues that could become long term issues (e.g., avoiding jobs that might lead to injury). But low income individuals don’t have the resources for basic health self-management and even simple interventions might have a big impact. My take home? Think about the chain and the closer you are to the end, the more focused the policy effects will be, if it exists at all.

50+ chapters of grad skool advice goodness: From Black Power/Grad Skool Rulz

* Brilliant Graduate Student

Written by fabiorojas

June 5, 2014 at 12:01 am

Posted in economics, fabio, research

Follow

Get every new post delivered to your Inbox.

Join 1,130 other followers