Results for “education signaling” 84 found
Saturday assorted links
2. Will Wilkinson on why America is polarizing.
3. “Economies governed by former economics students grow faster than economies governed by leaders with other education backgrounds; a result which is most evident for presidents.” Speculative, by the way.
4. Men differ more from women in personality in more developed cultures.
5. Harald Uhlig on North Korea.
6. Steve Francis practiced basketball shots with a high arc.
Tuesday assorted links
1. “Dolce & Gabbana used drones to carry handbags down the runway instead of models.”
2. The Uchida concert. It was remarkable how many people I know I bumped into there.
3. Bryan Caplan responds on education. I say businesses can simply use and encourage cheaper methods of signaling for potential employees. You don’t need “new and weird” systems, rather there is already considerable diversity within higher education and quantities can shift in the interests of economization. From another corner, Taleb reviews his reviewers.
4. What is mood?: a computational perspective.
5. How is the Sidewalk Labs Toronto waterfront project going?
Where do sheepskin effects come from?
From a loyal MR reader:
I’m very curious about the macroeconomics of the sheepskin effects. Traditional productivity forecast research tends to assume the wage premium is entirely human capital. Eg, Bosler/Daly/Fernald/Hobijn use a mincer equation with five education dummies http://www.frbsf.org/economic-research/files/wp2016-14.pdf Jorgensen’s approach dividing workers into types also assumes this is not an issue.
If sheepskin effects are purely relative status effects, then the impact on total output and income should be zero, right? This implies increasing educational attainment will have a much smaller impact on productivity and output than typical productivity forecasts imply.
But it seems to me like showing you are “high ability”, if that’s all it does, makes you able to be slotted into higher ability jobs, and that this won’t simply give you a leg up on other workers but increase the number of higher ability jobs filled.
Anyway, I’m sort of thinking out loud but would be curious to read a blog of your thoughts on this, so consider this a bleg!
In the simplest Spence signaling model, the output goes to workers, and if one more worker sends the signal and boosts his or her wage, the non-signaling workers will receive an equal amount less. That is an equilibrium condition, but it makes less sense as an account of dynamics. As a practical matter, it’s not clear why the employers should revise their opinion downwards for the marginal products of these less educated workers. You could say that competition makes them do it, but it’s tough to have good intuitions about an equilibrium that is hovering between/shifting across a varying degree of pooling and separating.
As a more general extension of Spence, a richer model will have market power and payments to capital and labor. If one more worker finishes school, that worker is paid more and the higher wage serves as a tax on production. Yet it is a tax the boss does not perceive directly. The boss thinks he is getting a better worker for the higher wage, but in the counterfactual with more weight on the pooling solution, the boss would have hired that same person, with the same marginal product, at a lower wage. The “whole act of production” will be and will feel more costly, including at the margin, but the boss won’t know how to allocate those costs to specific factors. By construction of the example, the boss however will think that the newly educated laborer is the one factor not to be blamed. So he’ll cut back on some of the other factors, such as labor and land. Labor in the company will be relatively more plentiful, and the marginal product of labor in that company will fall. So the incidence of a boost in the sheepskin effect falls on the land and capital that have to move elsewhere, plus to some extent the declining marginal products and thus wages for the remaining workers in the firm under consideration. Note that outside firms are receiving some influx of capital and land, and so in those firms the marginal product of labor and thus its wage will go up somewhat.
Or so it seems to me. The trick is to find some assumptions where the hovering between/moving across a varying degree of pooling vs. separation is not too confusing.
Friday assorted links
1. Arnold Kling’s books of the year. And Scott Sumner outlines his forthcoming book on the Great Depression.
2. “Arguably the least appreciated resource for Islamic State is its fertile farms.” (noisy video at that link, sorry, but the content is interesting)
3. Historical estimates of Chinese gdp. And where in the world is the greatest economic exposure to China?
4. Who says education is just signaling? Only bird brains?
5. “Some teams are accusing other teams of cheating, but there’s no cheating because there’s no rules,” (the culture that is curling).
Ten years after enrollment
At Bennington College in Vermont, over 48 percent of former students were earning less than $25,000 per year. A quarter were earning less than $10,600 per year. At Bard College in Annandale-on-Hudson, the median annual earnings were only $35,700. Results at the University of New Mexico were almost exactly the same.
There is more here from Kevin Carey. There is the well-known debate between human capital and signaling theories, but sometimes education is neither…
Is teaching about instruction or selection?
That is the title of a short essay by Gary Davis, here is the essay in toto:
Teaching is commonly associated with instruction, yet in evolution, immunology, and neuroscience, instructional theories are largely defunct.
We propose a co-immunity theory of teaching, where attempts by a teacher to alter student neuronal structure to accommodate cultural ideas and practices is sort of a reverse to the function of the immune system, which exists to preserve the physical self, while teaching episodes are designed to alter the mental self.
This is a theory of teaching that is based on the inter-subjective relationship between teacher and learner. This theory posits that teaching does not, as is commonly assumed, take place via instruction from teacher to students, but rather through a process of selection in the learner’s brain, stimulated by materials and activities utilized by the teacher. In this theory, the mechanism that drives the selection process in learners’ brains is co-regulated emotional signaling between teacher and learner. From this perspective, the power of formative assessment is that it intrinsically carries with it emotional aspects for both learner and teacher, in that it provides a feedback relationship between them both, and so, according to the Greenspan & Shanker theory of cognitive symbolic development, promotes cognitive development.
That is from the Journal of Brief Ideas, a new and worthy web site, and for the pointer to the site I thank Michelle Dawson.
Obama’s free community college plan
David Leonhardt writes:
The plan — which would require congressional approval — would apply to students attending a two-year college, including part time, so long as the college offered credits that could transfer to a four-year college or provided training that led to jobs.
David’s article is excellent and has much useful information:
As Reihan Salam of National Review notes, community college tuition is already low. In fact, it’s zero, on average, for lower-income families, after taking financial aid into account. Vox’s Libby Nelson wrote, “Community college tuition for poorer students is often entirely covered by the need-based Pell Grant.”
One potential implication is that by making community college universally free, the government is mostly reducing the cost for higher-income families.
Calculating the completion rate at community colleges is difficult, this estimate does some work to get it up to 38 percent. What would the completion rate be for the marginal students encouraged under the Obama plan? We don’t know, but I’ll guess at 20-30%, no more. That’s the real problem.
Furthermore some of the value of education is signaling to the labor market that you are able to finish college. I do think the learning component of education is generally more important, but for “marginally not attending community college individuals” — who are often regarded with suspicion by employers — I would not be surprised if the signaling component were one third or more of the value of a degree. To that extent, pushing more marginals into the degree funnel lowers the value of the degree for the others who were getting it already by lowering the average productivity of the pool of finishers. That would lower the efficiency gains from the program and also partially offset some of the intended distributional consequences.
Mike Konczal likes the idea, and believes it may lower higher education prices more generally. Libby Nelson at Vox considers it to be a middle class benefit. Neil McCluskey at Cato is negative. Carrie Sheffield is critical. Here is a look at potential winners and losers in the higher education sector. The plan could lead to federal money replacing state money, rather than leveraging it.
Citing the growing economy and improving labor market, Andrew Flowers noted:
…college enrollment is declining for recent high school graduates (those 16 to 24 years old). And it’s falling fastest for community colleges.
Overall my take is that the significant gains are to be had at the family level and at the primary education level, and that the price of community college is not a major bottleneck under the status quo.
Assorted links
1. Will you hire someone to do your grocery shopping for you?
2. New DSGE blog.
3. The Philadelphia Eagles are now targeting college graduates. Educational signaling comes to NFL football.
4. Can conservatives be funny?
5. Bill Gates on Jeffrey Sachs.
Assorted links
1. Signaling, part I, and signaling part II. And reminiscences of former heroin users.
2. Arthur Chu on the science of winning Jeopardy.
4. Predicting medals at the Winter Olympics.
5. Which out of print books do people search for the most?
6. You, Sir, are a fish face. And some species of crocodiles can climb trees. And my instinct is always to side with Robert Trivers.
The rise of West German wage inequality
This paper (pdf) by Card, Heining, and Kline appeared earlier this year and is published in the QJE and somehow it escaped my notice. Here is the first part of the abstract:
We study the role of establishment-specific wage premiums in generating recent increases in West German wage inequality. Models with additive fixed effects for workers and establishments are fit in four sub-intervals spanning the period from 1985 to 2009. We show that these models provide a good approximation to the wage structure and can explain nearly all of the dramatic rise in West German wage inequality.
…two-thirds of the increase in the pay gap between higher and lower-educated workers is attributable to a widening in the average workplace pay premiums received by different education groups. Increasing workplace heterogeneity and rising assortativeness between high-wage workers and high-wage firms likewise explain over 60% of the growth in inequality across occupations and industries. Finally, we investigate two potential channels for the rise in workplace-specific wage premiums: establishment age and collective bargaining status. Classifying establishments by entry year, we find a trend toward increasing heterogeneity among establishments that entered the labor market after the mid-1990s, coupled with relatively small changes in the dispersion of the premiums paid by continuing establishments. The relative inequality among newer establishments is linked to their collective bargaining status: an increasing share of these establishments have opted out of the traditional sectoral contracting system and pay relatively low wages.
I would put it this way: there are “high human capital firms” and “low human capital firms” to a greater extent than before. This change represents increasing German wage inequality fairly well, although it cannot be inferred that this increasing segregation causes the inequality increase.
Since newer firms reflect higher inequality and higher segregation, and I don’t foresee a major return of unionization, I take this as prima facie evidence that wage inequality in Germany (and many other places) is likely to continue to rise. This is another example of “the great reset,” as the newer firms offer a greater glimpse into the German economic future.
By the way, it is papers like this which increase my skepticism about the signaling model of education. The relevant signals being fed into these markets haven’t changed that much. Wage patterns are changing a lot.
For the pointer I thank Christian Odendahl.
Bryan’s breakdown on my break down
Re: my post immediately below on signaling, Bryan thinks I haven’t answered his question about the importance of signalling. But I have, he is just confused because I don’t use the exact same normalization as he does. In any case, I postulate the wage return to signalling as going away within five years, in say a career of forty years, then with the measure adjusted for the presence of capital and resource income. You can express that in terms of totals, variations, percentages, as you wish but the point remains that signaling is only a temporary factor, and overall only somewhat of a marginal factor (5-10%?) in explaining the overall evolution of wages. That is why it has lost ground to human capital approaches, all the more so with increasing inequality. (One can believe all of that and still think, as I do, that we could organize current education more efficiently and at lower cost.) I also stand by my points that insisting on the break down is missing the more important points about indeterminacy and nestedness and those points too can be applied to any normalization of the units. It would be more useful if Bryan would outline where he disagrees with my assessment, as the entire chain of reasoning is laid out pretty explicitly.
By the way, the easiest way to boost the contribution of signaling is to invoke the “you got your first job by signaling and then from that job quickly gained persistent extra human capital” argument, but even then that increment can, under traditional measures, be assigned to human capital. (You don’t want to rule out all human capital influences, on the grounds that signaling helped create them, any more than you wish to classify gains from signaling as human capital, if the human capital helped you get into the position to signal. But if that doesn’t convince you, revisit the earlier point about indeterminacy, as you can see that the marginal products for human capital and signaling will sum to well over one hundred percent.)
Agricultural yields and the returns to schooling
Agricultural yields, which reflect real returns and are not contaminated by “signaling,” are another independent way to measure the returns from schooling. Starting with Ted Schultz, this is a significant theme in development economics, and now from John Parman we have a new paper on schooling and agriculture in early 20th century America:
Formal schooling has a significant impact on modern agricultural productivity but there is little evidence quantifying the historical importance of schools in the early development of the American agricultural sector. I present new data from the Midwest at the start of the twentieth century showing that the emerging public schools were helping farmers successfully adapt to a variety of agricultural innovations. I use a unique dataset of farmers containing detailed geographical information to estimate both the private returns to schooling and human capital spillovers across neighboring farms. The results indicate that public schools contributed substantially to agricultural productivity at the turn of the century and that a large portion of this contribution came through human capital spillovers. These findings offer new insights into why the Midwest was a leader in the expansion of secondary education.
The paper is here, hat tip goes to the always excellent Kevin Lewis. Excellent ungated slides you will find here. One of the early slides indicates that in stable conditions “experience” outperforms education for generating agricultural productivity, but the value of education is high during times of dynamic change.
Why is there a shortage of talent in IT sectors and the like?
There have been some good posts on this lately, for instance asking why the wage simply doesn’t clear the market, why don’t firms train more workers, and so on (my apologies as I have lost track of those posts, so no links). The excellent Isaac Sorkin emails me with a link to this paper, Superstars and Mediocrities: Market Failures in the Discovery of Talent (pdf), by Marko Terviö, here is the abstract:
A basic problem facing most labor markets is that workers can neither commit to long-term wage contracts nor can they self fi nance the costs of production. I study the effects of these imperfections when talent is industry-specifi c, it can only be revealed on the job, and once learned becomes public information. I show that fi rms bid excessively for the pool of incumbent workers at the expense of trying out new talent. The workforce is then plagued with an unfavorable selection of individuals: there are too many mediocre workers, whose talent is not high enough to justify them crowding out novice workers with lower expected talent but with more upside potential. The result is an inefficiently low level of output coupled with higher wages for known high talents. This problem is most severe where information about talent is initially very imprecise and the complementary costs of production are high. I argue that high incomes in professions such as entertainment, management, and entrepreneurship, may be explained by the nature of the talent revelation process, rather than by an underlying scarcity of talent.
This result relates also to J.C.’s query about talent sorting, the signaling model of education, CEO pay, and many other results under recent discussion. If it matters to you, this paper was published in the Review of Economic Studies. I’m not sure that a theorist would consider this a “theory paper” but to me it is, and it is one of the most interesting theory papers I have seen in years.
Failing versus Forgetting
Bryan Caplan has a very good post on the human capital and signalling models of education. The key point is this, under the human capital model someone who forgets knowledge is no better than someone who failed to learn the same knowledge. Under the signaling model, however, failing and forgetting are very different. Bryan illustrates:
If I’d failed Spanish, I couldn’t have gone to a good college, wouldn’t have gotten into Princeton’s Ph.D. program, and probably wouldn’t be a professor. But since I’ve merely forgotten my Spanish, I’m sitting in my professorial office, loving life.
Natural experiments and the return to schooling
Cowen’s First Law: There is a literature on everything.
Responding to queries from Kling and Caplan and Henderson, let us turn the microphone over to Andrew Leigh and Chris Ryan:
How much do returns to education differ across different natural experiment methods? To test this, we estimate the rate of return to schooling in Australia using two different instruments for schooling: month of birth and changes in compulsory schooling laws. With annual pre-tax income as our measure of income, we find that the naıve ordinary least squares (OLS) returns to an additional year of schooling is 13%. The month of birth IV approach gives an 8% rate of return to schooling, while using changes in compulsory schooling laws as an IV produces a 12% rate of return. We then compare our results with a third natural experiment: studies of Australian twins that have been conducted by other researchers. While these studies have tended to estimate a lower return to education than ours, we believe that this is primarily due to the better measurement of income and schooling in our data set. Australian twins studies are consistent with our findings insofar as they find little evidence of ability bias in the OLS rate of return to schooling. Together, the estimates suggest that between one-tenth and two-fifths of the OLS return to schooling is due to ability bias. The rate of return to education in Australia, corrected for ability bias, is around 10%, which is similar to the rate in Britain, Canada, the Netherlands, Norway and the United States.
There are many other papers in this genre, such as by Joshua Angrist, and they yield broadly similar results. Here is an Esther Duflo paper on Indonesia. There is an excellent David Card survey on the causal returns to education, from 1999, but more recent results have shown the same. Card’s conclusion:
Consistent with earlier surveys of the literature, I conclude that the average (or average marginal) return to education is not much below the estimate that emerges from a standard human capital earnings function fit by OLS. Evidence from the latest studies of identical twins suggests a small upward “ability” bias – on the order of 10%. A consistent finding among studies using instrumental variables based on institutional changes in the education system is that the estimated returns to schooling are 20-40% above the corresponding OLS estimates.
That last sentence is because the marginal student is especially in need of education. The view that education is mostly about signaling is inconsistent with the established consensus on the returns to schooling and yet the writers at EconLog do not respond to this literature or, as far as I can tell, even acknowledge it.
Here is one of my earlier posts on education. Here is my theory of (some) education.