Working and academic performance at university

Does working negative impact on students' academic performance at university? It's an interesting and important question, especially as students are increasingly finding themselves in financial positions where working is a necessity in order to meet their living costs, rather than a source of 'beer money' as it was in earlier decades.

However, the theoretical relationship between working and academic performance is not straightforward. As Moris Triventi (European University Institute) outlines in this 2014 article published in the journal Economics of Education Review (ungated earlier version here), there are four possibilities for this relationship:

1. Zero-sum: Every hour spent working is one less hour that could be spent studying or attending classes, and since studying and attendance are important for academic performance, working has a negative causal impact on academic performance;
2. Reconciliation: Students are reasonably flexible in their allocation of time between working, studying, and leisure, so an additional hour spent studying does not necessarily reduce studying or attendance, and students can be strategic in choosing classes that are less demanding, so there is likely no effect of working on academic performance;
3. Negative selection: Students who choose to work during university are different in meaningful ways from students who do not work, perhaps because they did not achieve as well at high school, or they don't qualify for scholarships or other assistance, or because their aren't as motivated towards their studies, so working is negatively correlated with academic performance, but the relationship is not causal;
4. Positive selection: Students who choose to work during university are high achievers and highly motivated, and are able to work because they are confident that it will not negatively impact on their studies, so working is positively correlated with academic performance or there may be no effect, but if there is a positive relationship then it is not causal.

Triventi goes on to test these theories using representative survey data collected from 1834 Italian freshman students in 2004. First, he separates students into three groups (non-working students; low-intensity workers; and high-intensity workers), and notes that:

Among Italian freshmen in 2002–2003, 17% worked up to 20 h during their first academic year, while 10.5% worked on average more the 20 h per week... Low-intensity workers worked on average 11.3 h per week, while high-intensity workers worked 35.4 h.

Then, looking at the relationship between working and academic performance (measured by the number of credits earned in the first year of university), he finds that:

Looking at high-intensity work there are few doubts: in all models it has a large detrimental effect on academic progression, even when controlling for both observed and unobserved variables. This means that the zero-sum approach fully applies to the condition of high-intensity workers, who devoted on average 35 h per week to their job. It is likely that such a degree of involvement makes it difficult to devote a sufficient amount of hours to study, and thus to maintain a regular academic progression...

The second noteworthy finding refers to the low-intensity employment status. In this case, the results change in different models. Bivariate and traditional multivariate analyses show no major gap in academic progression between low-intensity workers and non-working students... once accounting for unobserved variables – which are likely to capture motivation and multitasking skills in our work – the picture changes, since the low-intensity employment status also negatively affects the number of credits acquired (albeit to a less extent compared to the high-intensity employment status). This means that, in line with the positive-selection hypothesis, the standard analyses in Italy mask the fact that low-intensity workers are positively selected (conditional on observed covariates) and thus are able to compensate the difficulties of being employed while studying by their higher commitment to pursue the two activities at the same time.

In other words, working does have a negative effect on academic performance. For students trying to study while working essentially full-time, the effect is negative and large (these students do much worse than non-working students). Students working part-time are able to compensate for the negative effects of working, but that is because of the types of students who decide to work part-time while studying.

The results of this study are interesting, and probably should worry us. As we increasingly move university teaching to more online and flexible modes, this opens up the possibility for more full-time workers to engage in studying, or for students who would otherwise be working a little or not at all, to work more. Both of those are likely to lead to a reduction in academic performance.

A university system that encourages students to take on paid work due to financial reasons is setting those students up to fail. To avoid this negative outcome, we really need a student allowances and loans system that is able to provide adequately for students' living costs (unlike what we have now).

How has applied microeconomics changed over time?

In a 2020 NBER Working Paper (alternative ungated version here), Janet Currie, Henrik Kleven, and Esmée Zwiers (all Princeton University) looked at how the methods employed in applied microeconomics research have changed over time. They mined text-based data from a sample of 10,324 NBER Working Papers over the period from 1980 to 2018, and 2,830 journal articles published in the top five economics journals from 2004 to 2019. They find a number of interesting trends, including:

...a virtually linear rise in the fraction of papers, in both the NBER and top-five series, which make explicit reference to identification. This fraction has risen from around 4 percent to 50 percent of papers...

...a somewhat slower rise in the use of experimental and quasi-experimental methods... Currently, over 40 percent of NBER papers and about 35 percent of top-five papers make reference to randomized controlled trials (RCTs), lab experiments, difference-in-differences, regression discontinuity, event studies, or bunching...

...a very similar pattern in references to administrative data. The NBER series starts increasing in the mid-1990s, rising to about 30 percent today. The top-five series shows a similar increase, but with a lag of about three years... The term Big Data suddenly skyrockets after 2012, with a more recent uptick in the top five...

The importance of figures relative to tables has increased substantially over time and in two phases. The first phase happened in the 1990s and likely reflects the diffusion of new software such as STATA that made it easier to create impactful figures. The second phase has happened in the last 10-12 years and is still ongoing. This is likely due to the increasing use of administrative datasets, which lend themselves to compelling graphical representation using raw data and non-parametric approaches...

...a sharp rise in the fraction of NBER working papers discussing randomized controlled trials since 2005, and especially since 2010...

Laboratory experiments have grown steadily in popularity since the late 1990s, which is connected to the rise of Behavioral Economics during this time period...

...authors have become increasingly concerned with whether their estimates are precisely estimated, and not merely with whether they are significantly different from zero in a statistical sense...

...a sharp rise in references to confidence intervals since the mid-1990s...

...after year 2000, there has been a massive increase in attention paid to clustering of standard errors...

Currie have clearly identified the most important changes in the types of research methods used in applied microeconomics over the last 30 or so years. One thing they haven't noted is the rise in the use of textual analysis, including sentiment analysis, as a trend within the use of big data. This is somewhat ironic, since that's what their paper uses!

It would be interesting to see a similar exercise conducted for applied macroeconomics.

[HT: Marginal Revolution, last year]

Scalpers and the irrationality of Ontario Parks

CBC reported earlier this week:

Ontario Parks is cracking down on people who book camping sites and resell their reservations for profit.

The province doesn't condone reselling reservations. because it's been particularly difficult to book a site, said a spokesperson for Jeff Yurek, minister of environment, conservation and parks.

This year has seen a particular surge in campsite bookings and competition for coveted spots...

"We know that there are instances where individuals are attempting to sell reservations with the intention to make a profit," Chelsea Dolan said in an email to CBC Kitchener-Waterloo on Thursday evening.

As of Saturday, anyone with reservations won't be allowed to resell them.

A common rationale for preventing the scalping (or resale) of tickets is that the scalpers make consumers worse off. But should that in itself be a reason to discourage scalping? Consider the market for camping sites, where there is a fixed number of sites made available, as shown in the diagram below. The supply of camping sites S0 is fixed at Q0 - if the price rises, more sites will not suddenly be made available (note that the diagram assumes that the marginal cost of providing sites up to Q0 is zero).

Demand for camping sites is high (D0), leading to a relatively high equilibrium price (P0). However, camping sites are priced at P1, below the equilibrium (and market-clearing) price. At this lower price, there is excess demand for camping sites (a shortage) - the quantity of sites demanded is Qd, while the quantity of sites that are available is fixed at Q0.

With the low camping site price P1, the consumer surplus (the difference between the price the consumers are willing to pay, and the price they actually pay) is the area ABCP1. Producer surplus (essentially the profits for Ontario Parks) is the area P1CDO. Total welfare (the sum of producer and consumer surplus) is the area ABCDO. At the higher price P0 due to the actions of scalpers (buying at P1 and selling at P0), the consumer surplus decreases to ABP0, while producer surplus remains unchanged. The scalpers gain a surplus (or profit) of the area P0BCP1, and total welfare (the sum of producer and consumer surplus, and scalper surplus) remains ABCDO. So the camping site scalpers don't change total welfare at all, just the distribution of that welfare between the parties. However, they do clearly make consumers worse off as a group, because consumer surplus is lower.

However, let's go back to considering the price at P1. At that price, the excess demand means that there are many campers who are willing and able to pay a price that is higher than P1, who miss out on a camping site because of the excess demand. If the price was a little higher than P1, then that would reduce the amount of excess demand, and fewer willing buyers (at the higher price) would be missing out on a camping site. In fact, you would need to raise the price all the way to P0 before you get to a situation where no willing buyers are missing out. And, that's exactly what the scalpers do. By banning scalpers, Ontario Parks are essentially protecting the excess demand. They are ensuring that some people, who are willing and able to pay the market price, will miss out on camping sites.

You might argue that raising the price will squeeze consumers out of the market, starting with those who are willing and able to pay the least for a camping site, and that such an outcome is unfair. [*] But is it really fairer that there are many people willing to pay the market price who are missing out on camping sites?

Finally, Ontario Parks is clearly missing a trick here. Why are they pricing so low, and allowing scalpers to earn a surplus at all. If Ontario Parks raised the price of camping sites to P0, there would be no market for scalpers, or at least there would be no profits for them to make, and all of the combined producer surplus and scalper surplus would belong to Ontario Parks. Think of how much of an improved service Ontario Parks could offer with all that additional revenue. What are they thinking?

[HT: Marginal Revolution]

*****

[*] Market pricing is incredibly unfair. Personally, I strongly believe that the pricing of Lamborghinis is unfair. They should be priced in such a way that I can afford to drive a different coloured one to work each day of the week. I demand action!

Read more:

• The Super Rugby final and ticket scalping
• Harry Potter and the chamber of scalpers
• The economics of ticket scalping

Meaningful jobs and compensating differentials

In The Conversation last week, Andrew Bryce (University of Sheffield) wrote about the meaningfulness of jobs. These bits in particular caught my eye:

Work provides many things over and above the monthly pay cheque: status and identity, community and social connection, doing tasks that we find stimulating, and the opportunity to make a positive contribution to society. All of these things make work feel meaningful.

My research explores how paid work is experienced as meaningful compared to the other activities people do in their everyday lives. I also identify the types of job in which people experience the most meaningfulness and explore how these results can be explained by the particular qualities of different occupations...

People in community and social service occupations (which includes social workers, counsellors and clergy) experience the most meaningfulness in their work.

The other top-ranking occupations are: healthcare practitioner and technical occupations; education, training and library occupations; and, perhaps surprisingly to some, legal occupations. More broadly, people working in the non-profit sector and self-employed people report significantly more meaningfulness in their work than those employed in private sector for-profit firms...

When work is meaningful, then that becomes a reward in itself and generous pay offers are not prioritised to motivate people and retain staff. In contrast, less meaningful work has no such intrinsic value, so a monetary reward is needed to get people to do these jobs.

This of course leads to the perverse situation where the most socially useful jobs are those that are paid the least. It may seem unfair but it’s the reality of how the labour market works.

It's the reality of how the labour market works because of compensating differentials. Jobs have both monetary and non-monetary characteristics. Monetary characteristics include the pay and other monetary benefits. Non-monetary characteristics include the whole range of other things associated with the job. Perhaps it is dirty, dangerous, or boring. Or perhaps it is clean, safe, or fun. When a job has attractive non-monetary characteristics, then more people will be willing to do that job. This leads to a higher supply of labour for that job, which leads to lower equilibrium wages. In contrast, when a job has negative non-monetary characteristics, then fewer people will be willing to do that job. This leads to a lower supply of labour for that job, which leads to higher equilibrium wages. It is the difference in wages between jobs with attractive non-monetary characteristics and jobs with negative non-monetary characteristics that we refer to as a compensating differential (essentially, workers are being compensated for taking on jobs with negative non-monetary characteristics, through higher wages).

The meaningfulness of a job is a non-monetary characteristic. If a job is meaningful, then more people will want to do the job (holding other job characteristics constant), and wages will consequently be lower. Another way of thinking about it is that employers of workers who offer meaningful jobs don't have to compete hard to attract workers, and so they don't have to offer as high a wage to fill the job. Either way, meaningful jobs pay less because of compensating differentials.