Monthly Archives: December 2013

Scranton, again, again

As a follow-up to my follow up.

How hard would it be to incorporate the fading of old, low-cost technology into a multi-generation macro model? Romer has been parameterizing the macroeconomy with technological change since the ’80s. It doesn’t sound like such a big logical jump to incorporate the Jencks model.

Advertisements

Leave a comment

Filed under Uncategorized

Database of Data Sets

I would love to see a comprehensive database of data sets that is searchable by variable name, frequency, survey methodology, target population, etc. ICPSR is okay for finding data, but doesn’t do a good job of putting that data in context and doesn’t facilitate comparisons between similar-but-not-the-same data sets. The fact that notes like this have to exist is a sign that decentralized access to data poses an obstacle to research.

It doesn’t have to be like the St Louis FRED or IPUMS, where you can actually access the data. All it needs is a link to wherever the data is hosted, or contact info for the researchers who developed it in the case of one-off projects. Maybe even a link to the corresponding ICPSR page. What it does need is an exhaustive list of available data items in each set and a standard format for every summary entry. Look at the Wikipedia page for any  country. On the right side of the page is a sidebar with major relevant features, and most sidebars have roughly the same structure. How big is the panel? What’s the attrition rate? What’s the survey frequency? Right there, in your face.

The biggest problem I have with the search tools currently available is that the quantitative  specifics of the data (i.e. the most important part) are never plainly available and are buried deep in the data file. The right data is always out there, but finding it is unacceptably laborious. I don’t want to have to spend a half hour that horrible DataFerett app only to find out that SIPP doesn’t ask people if they received vocational training. I don’t want to have to rely on NORC to post a chart of which variables are available in which years, because who knows where they posted it? I don’t want to have to pore over original survey questionnaires just to find out if the variable I need is even in there. The BLS tries admirably to keep their employment data sets organized, but why should we make them bend over backwards to do it? They even give me SAS, SPSS, and Stata files for importing and recoding data from the NLS Investigator– daiyenu, shouldn’t that be enough?

I should be able to type something like “real income” into a search box and immediately get a list of every data set that records real income by whatever unit of analysis. I should be able to specify date ranges and broad classifying categories. Let’s say the CPS comes up, and I click on it. Immediately I see the sample size, the frequency, the target population, who produces it,  a few links to very similar data (think “related artists” on Spotify), a link to where the CPS data can be downloaded, and a brief description of the sample design. Maybe even a link to Scopus or Web of Knowledge with a search for “Current Population Survey” already punched in. Because I searched for “real income,” variables with description and tag matches appear first. I can click on “See All Variables” and have instant, searchable access to the full list of every variable available in every year. I do not need summary statistics, I do not need an FTP link, I do not need SAS code. I just want to know what’s in the damn data set, and I don’t want to have to actually download the data to find it.

Everyone on the internet uses “tags” for everything (except me because I’m lazy). How stupidly easy is it to search StackExchange, Reddit, and Last.FM? Very, and it’s because people tag stuff. This databse could add a tag to each variable in the case where two similar variables don’t have enough common words in their descriptions to come up together in searches. Tag different measurements of the same concept with the same tag: instant side-by-side comparison capability. Better yet, actually be able to query variables from different data sets and run a side-by-side comparison the way you can online with  computers and cars.

This is a big project, but if enough effort goes into building the structure very little effort will be needed to maintain it.

UPDATE: Someone in the government made a valiant attempt with http://www.data.gov/, but it’s buggy and none of the data download links seem to work. They went way too big way too soon.

UPDATE 2: I think SIPP actually does ask people if they received vocational training, but it’s buried somewhere in a many-dozen page PDF file in a part of the website I will never be able to find again.

Leave a comment

Filed under Uncategorized

SBA and ETA… and behavioral nutrition

The Small Business Administration requested a budget of about $1.2b in 2013. The Employment and Training Administration requested a budget of about $3.2b.

To study:

  1. Which administration meets its own goals better?
  2. Which administration contributes more to economic welfare?
  3. Which administration creates more employment? Which administration creates better employment?
  4. Which administration actually helps more people, which people, and how?
  5. Why don’t we have a (possibly government-funded) microfinance system in the US?*

*Better question: why don’t we have a microfinance system in the US while India has an extensive one? Poor Economics (Banerjee & Duflo, 2012) makes a case for and against microfinance, and what I took away from the microfinance chapter doesn’t seem like it couldn’t apply to the much-less-but-still-very poor. They discuss in great detail how the very poor behave, and yet don’t do a great job of distinguishing the behavior of the absolute poor (<$1 a day in a corrugated steel hut) from that of the relative poor (<$18,000 a year in a project).

And somewhat unrelated: how big a role does the “effort exhaustion” effect in Thinking Fast and Slow (Kahneman, 2011) play in poor diet and lifestyle choices by the working poor and people in general? Can the negative outcomes of trying to help people by telling them what to do (“you shouldn’t drink soda, it’s bad for you and you are stupid” versus “check out how good fresh squeezed OJ tastes”) have a macro-level effect?

1 Comment

Filed under Uncategorized

Scranton, again

I have an NBER working paper to back me up on my Scranton post. The authors emphasize a “rising level of skill and skill premium,” and therefore higher relative wages for skilled workers.

That’s great, but what the purely economic argument misses is that this is not Pareto optimal. If one group benefits and another stagnates, the stagnating group will eventually be pushed to the fringes of society. Worse, the goods and services they depended on could become unavailable if their costs rise. There’s a better and more thorough explanation of this effect in Jencks, et al.

I would love to see a model that produces this effect endogenously.

1 Comment

Filed under Uncategorized

Fuzzy type in game theory

Last post I said “I think there are a lot of workers that fall into this category at least partially.”

Typing and classification in social sciences is often not mutually exclusive, and is rarely definitive. If it were, we wouldn’t have hierarchical clustering, machine learning, and regression models with categorical response variables. I’d like to know how well game theory, auction theory, and other type-dependent models generalize to fuzzy typing, even with a discretized scale of inclusion (i.e. “fully type 1”, “partially type 1”, “not at all type 1” instead of a [0,1] scale).

So naturally I searched something like “fuzzy type game theory” in Google, and found that a Mathematics MSc student at Iowa State wrote a thesis on the subject. There’s also a paper written by an economist and a mathematician at Louisiana State with a promising title.

Leave a comment

Filed under Uncategorized

Market power is monotone with the size of the choice set: Implications

The reasoning in the previous post implies that an individual worker’s market power is probably lower than what we’d assume or estimate (however you’d operationalize market power) using homogeneous human capital. Caveat for what follows: I’m not a wizard with game theory and I know very little about bargaining in particular.

Roberts said that a worker’s market power derives from the size of that worker’s choice set. Let’s operationalize market power as “the ability to willfully influence one’s own terms of employment.” I can think of two jointly sufficient conditions for a worker to have any market power at all:

  1. the worker can credibly threaten to defect from wage bargaining,
  2. the employer would rather continue to bargain than let the worker defect.

It might be hard to prove, but I suspect that these conditions are also necessary. If they are, then workers’ market power might not be that large in general. Net of social institutions, how much can a typical worker ask of an employer? This framework could probably be generalized to any bargaining situation, and I’d be shocked if nobody has written about it.

One interesting “prediction” it generates: consider a type of workers who are afraid of losing their jobs, and who employers know which workers are of this type. How credible is their threat of defection? How much market power do they have?

I think there are a lot of workers that fall into this category at least partially.

Leave a comment

Filed under Uncategorized

Market power is monotone with the size of the choice set

Inspired by this old blog post, particularly this quote:

What keeps my wages high (and yours) is our alternatives. Is there any evidence that workers have fewer alternatives than they once had? If anything, workers are more mobile today than ever. What sets workers wages are the wages of those alternatives. That depends, generally, on our skills, our knowledge and our drive and discipline—human capital and how well we are able to apply it. Workers are better educated than ever. That is why I believe that compensation, properly measured, is higher than it was five or ten or twenty or thirty years ago.

Russ Roberts really captures the essence of market power here, and in doing so captures the essence (better than I ever could have) of what I think is missing in the current debate over skills-biased technological change (SBTC).

I’m not convinced that workers’ alternatives are generally as numerous as Roberts claims. Moreover, I’m not sure that the availability of job choice sets, partially ordered by set containment, is supermodular as a function of human capital; that is, increasing the absolute quantity of human capital might not increase the size of the choice set as Roberts suggests. Two unignorable cases are overqualification and job-specificity of skills.

Skill-biased technological change could be increasing skills specificity, as laid out in Unchecked and Unbalanced by Arnold Kling. Without some countervailing force in the economy, this could actually shrink individuals’ choice sets, reducing their market power even as their human capital grows. I believe this mechanism has a role to play in the recent concerns of income inequality and I have not heard of any studies on it.

Roberts wrote his post before the Great Recession, but people are amazingly still comfortable talking like this in 2013. To me, it is inexcusable to ignore this mechanism when studying SBTC and related models.

edit: I said nothing in this post about the level of wages in one’s alternative set. I realize now that I was more properly referring to the subset of the alternative set that is weakly preferred to the current job.

Leave a comment

Filed under Uncategorized