Relative prices and parental leave

The New Zealand Herald reported earlier this week:

When Sammy Phillipson was planning to start a family with his wife, taking three months off to care for the babies was never on the table.

“It just wasn’t a financial opportunity for our family,” the national business manager says, lamenting on the costs of their Auckland mortgage.

As his wife, Naomi Williams, was to be the primary caregiver, under New Zealand law Phillipson was only entitled to two weeks of unpaid partner’s leave, at a time when the family’s costs would be rising.

But in October 2021, when his first child, Margot, was 15 months old, his employer, beverage company Lion, implemented a new policy allowing all parents to have 12 weeks of paid time off following the birth of their children.

Maternity leave is technically for the first year of the baby’s life, but carers at Lion are able to take the 12 weeks anytime in the first two years of their child’s life.

Lion had been offering the paid leave to the primary caregiver for more than three years. This was on top of the Government’s parental leave allowance, which was upped to six months from July 2020.

But since Lion’s extension to all new parents, the company says it is now seeing a 50/50 split of men and women taking parental leave.

When the relative price of something decreases, then people tend to do more of it. When the relative price decreases by a lot, then people will tend to do a lot more of it.

An opportunity cost is the value of an activity, measured in terms of the next best alternative foregone. In this case, the opportunity cost of parental leave is the wages foregone during any unpaid leave. Taking parental leave that is unpaid (or mostly unpaid) therefore comes with a relatively high opportunity cost. In that case, we can say that the relative price of parental leave is high.

On the other hand, when unpaid leave is replaced with paid leave, the opportunity cost of the parental leave decreases. The relative price of parental leave decreases, and we should expect people to take more parental leave.

From the article:

Since the leave was implemented for all parents, [Lion New Zealand people and culture director Jacquie Shuker] said the outcomes have exceeded their expectations in terms of men taking up the offer.

The generosity of the parental leave provisions obviously matters. The change from unpaid to paid leave is a substantial decrease in relative price. I'm not at all surprised that there is a big response to it. And neither would I be surprised by large responses to any of these (from the same article):

Lion is not the only company offering additional help for new parents. Financial services firm EY is another with extra support - from next month employees with any service period get 26 weeks of paid parental leave, which can be used flexibly and during the child’s first 24 months.

In November, Contact Energy announced it would offer primary caregivers a full salary top-up for the 26 weeks Government parental leave period, 3 per cent KiwiSaver for the duration of the worker’s parental leave and six months of flexible working, meaning employees can choose to work 80 per cent of their normal weekly hours but still receive full pay for their first six months.

It would also give primary carers $5000 towards childcare, 10 days special leave for pregnancy-related appointments, three months free power for employees who are also customers and a food package with pre-prepared meals on the arrival of the baby.

Employees who become parents but are not the primary carer are also offered four weeks of paid leave, which can be taken over 13 months, three months of free power as well as the meals on baby’s arrival.

Vodafone announced last year it would top up the Government payment to full pay for 22 weeks, give primary carers an extra 26 days of paid leave and give partners 26 paid leave days that could be used flexibly over two years after the birth.

In August Z Energy said it would contribute 5 per cent towards KiwiSaver for all employees on parental leave for their entire parental leave period and pay employees working part-time (more than 20 hours a week) 5 per cent towards their KiwiSaver based on their full-time salary equivalent rather than their actual pro-rate pay.

In 2018, NZME, the Herald on Sunday’s parent company, started offering a $5000 one-off payment to permanent employees who are primary carers, when they start parental leave. This is the equivalent of an additional nine weeks of paid leave.

NZME also offers two weeks of paid leave to partners.

It would be interesting to know whether more modest increases in employers' parental leave provisions also lead to increased uptake. But it is not surprising at all that large changes lead to large responses. When the relative price of something decreases, then people tend to do more of it. When the relative price decreases by a lot, then people will tend to do a lot more of it.

First-year-fees-free papers are not free if you fail

The New Zealand Government introduced free tertiary education (for the first year) in 2018. Since then, it's been interesting to note that the number of 'ghosts' in first-year classes (like my ECONS101 and ECONS102 classes) has increased markedly. These 'ghosts' enrol in the class, and then never show up for any lectures or tutorials, and don't complete any assessment at all. It's hard to say what enrolling actually achieves for them - perhaps a year of student allowance?

Anyway, when talking with other ('non-ghost') students about this situation, many of them mention the first year of study being free of fees as a reason. The argument is that failing a paper has no cost at all, because the paper is free of fees. In this short post, I'll explain why the idea that failing a free paper has no cost at all is quite wrong. At least, that it is wrong for any student who is aiming to complete a qualification. [*]

Let's consider a standard three-year degree qualification. At the University of Waikato, this is made up of 360 points. Most papers are 15 points, so that means completing the degree takes 24 papers spread across three years. The fees-free policy covers the first 120 points, which for most students is eight papers. Let's say that ECONS101 is one of those eight 'free' papers.

If the student passes ECONS101 (and their other papers), then they end up paying for 16 papers over the course of completing their degree. However, if they fail ECONS101, then they have to take an additional paper (either taking ECONS101 again, or another paper in place of ECONS101). That means taking 25 papers (instead of 24), of which 17 cost the student real money (instead of 16). So, failing ECONS101 (or, indeed, any other paper during the fees-free year) is not free at all. It costs the student the full cost of a paper, since they now have to take one additional paper to complete their degree.

So, it should be easy to see that failing a paper in the first year is not free, even if the paper doesn't attract any fees the first time around. And we haven't even talked about the opportunity costs of completing the paper that the student failed, which is made up of the time and effort (if any) the student spent attempting the pass the paper the first time.

Now that we've dispelled the myth that failing papers is without cost, it's time to focus on studying, and improving the chances of avoiding any extra costs. Teaching starts this week for A Trimester!

*****

[*] For students who are not aiming to complete a qualification, then the monetary cost really is zero (since the policy covers both tuition fees and any mandatory student services fees). The only cost to a 'ghost' is the opportunity cost of the time it takes to complete their application to enrol. In that sense, the students are correct that fees-free may explain the increase in the number of 'ghosts'.

Book review: The Economics Book (Steven Medema)

The list of books that I would recommend that every student in economics read is very short [*]. However, I now have one new book to add to that list: The Economics Book by Steven Medema. I was first attracted to this book in the Unity Books store in Auckland. It had a shiny blue cover that appealed some inner magpie instinct, and the word 'economics' on the cover. And inside, lots of pictures.

Through the book, Medema takes the reader on a tour of the history of economics. The book is structured as a series of short essays, each just a page long and devoted to some aspect of the history of economic thought, a key publication, or a key moment in history that had economic implications. The book starts with Hesiod's Work and Days, and ends with cryptocurrency (the book was published in 2019). Along the way, Medema writes about tulipmania, Mandeville's Fable of the Bees, several pages devoted to Adam Smith's writings, the Austrian School, the 'circular flow diagram', the permanent income hypothesis, agency theory, endogenous growth theory, and much, much more. I made around two pages of short notes on things to incorporate in my teaching (as my students know, I often like to namecheck the economists who were the origins for the theories we cover in class).

I also learned a lot from this book. One lowlight of my own economics education was a lack of exposure to the history of economic thought and key moments in the development of economics as a discipline. This book fills the gap nicely, and it is for that reason that I think it should be required reading for economics students. But not just for economics students. This is a book that an economics student could give to their parents, to help them understand a bit more about economics (unless, of course, the parents are also economists!).

Medema writes in an interesting and engaging style, making each short essay a pleasure to read. He also handily cross-references the essays, so that readers can follow particular lines of thought, forwards or backwards through the book. The pictures that accompany each essay are added value, although this is not quite a coffee table book.

There were several stand-out essays for me, especially early in the book. I hadn't heard of Aristotle's writings on justice in exchange, which Medema illustrates with Aristotle's tale of an exchange between a cobbler and a builder:

Each man receives something of value from the other, and each incurs costs in producing the goods offered. For an exchange to be just, Aristotle posits that the ratios of benefit to cost for each party must be equal. A house provides great benefit to the cobbler, but the cost of producing a pair of shoes is very low. On the other hand, a pair of shoes for the builder provides a relatively small benefit, while the cost of building a house is very high. Given this, the cobbler should provide the builder with many pairs of shoes in return for a house, to satisfy the reciprocity requirement.

There are also excellent essays on Ibn Khaldun as a forerunner to Adam Smith, and the Salamancans as the origin of the quantity theory of money as early as the 16th Century. I also really appreciated the references to early female economists, including Jane Marcet and Harriet Martineau as early popularisers of economic ideas, and Joan Robinson's contributions to understanding monopolistic competition. Medema also offers what I think is the clearest explanation of the Shapley value that I have ever read.

Inevitably, in a book like this, there will be key ideas and developments that do not make the final cut or are not considered at all. I was truly surprised, though, that there was no mention of the Mont Pelerin Society, or not as much devoted to the neoliberal turn in economic policy in the 1980s as I would have expected (those two are, of course, closely related).

Nevertheless, this is an excellent book, and well worth purchasing. Highly recommended!

*****

[*] The short list of books that I think every economics student should read (along with this one), includes Robert Frank's Falling Behind: How Rising Inequality Harms the Middle Class (which I discussed here), Ziliak and McCloskey's The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives (which I reviewed here), and Akerlof and Kranton's Identity Economics (which I reviewed here).

Hamilton Gardens' entry fee for tourists as a form of price discrimination

I missed this story when it was first reported, but Hamilton Gardens has made the decision to introduce a $10 entry fee once their new entry precinct opens at the end of the year. However, entry for Hamilton residents will remain free. As Stuff reported:

The days of free entry to Hamilton Gardens are numbered for visitors from out of town as work begins on a new precinct at the flagship attraction.

Tourists will pay $10 for entry to the themed gardens when the entry precinct is complete, possibly by the end of the year, while Hamilton residents and children under 16 will get in for free.

Work is set to start next month on the area, which is also intended to better open up the Gardens for visitors, making it easier to see what is on offer.

This is a great idea, and backed by solid economics. I've made the point before that, if tourist operators are trying to maximise profits, then tourists should be paying more for tourist activities than locals (see here, and here, and here). To see why, let's look at the example of Hamilton Gardens. As the Stuff article notes:

Hamilton & Waikato Tourism chief executive Nicola Greenwell said the Gardens are one of the region’s top offerings.

“They're incredibly important to our visitor offering, they are one of our hero products. So they do attract a large number of people to our city and to our region.”

She thinks visitors will take the $10 fee in their stride. “I think that the offering that the Hamilton Gardens has is world-class, and it is well worth an entry fee.”

Tourists will take the $10 fee in their stride, but not just because of the world-class nature of the gardens. We'll get to that a bit later. First, some theory.

Price discrimination occurs when a firm sells the same product to different customers for different prices, and where the difference in price doesn't arise from a difference in costs. In this case, access to Hamilton Gardens will be free for Hamilton residents, but cost $10 for tourists. That difference in price does not result from a difference in costs, because it costs Hamilton Gardens the same to host a visitor regardless of where the visitor is from.

For price discrimination to work, three conditions have to be met:

1. Different groups of customers (a group could be made up of one individual) who have different price elasticities of demand (different sensitivity to price changes);
2. The seller needs to be able to deduce which customers belong to which groups (so that they get charged the correct price); and
3. No transfers between the groups of customers (since the seller doesn't want the low-price group re-selling to the high-price group).

In the case of Hamilton Gardens, there are two groups of visitors (locals and tourists). Why does the price elasticity of demand differ between these two groups? Tourists have lower sensitivity to price (low price elasticity of demand) for a couple of reasons. First, for visitors there are few substitutes for visiting Hamilton Gardens (or other tourist attractions). Locals have lots of alternative activities (like staying at home watching Netflix). When there are fewer substitutes, the price elasticity of demand is lower. Second, tourists have usually also travelled further than locals, at higher cost, in order to get to Hamilton in the first place. So, the cost of entry into Hamilton Gardens is pretty small in the overall cost of their holiday. Whereas for locals, the cost of entry into Hamilton Gardens (if there was an entry fee) is essentially the entire cost of their visit. When the price is a smaller component of the total cost (as it is for tourists visiting Hamilton Gardens), the price elasticity of demand is lower. For both of these, tourists will be relatively insensitive to price compared with locals, and raising the price of entry for tourists isn't going to keep them away in great numbers.

That covers the first condition for price discrimination. For the second condition, Hamilton Gardens needs to be able to tell who the Hamilton residents are (because they get free entry) and who the tourists are. And for the third condition, Hamilton Gardens needs to ensure that Hamilton residents can't 'buy' free tickets, and then give them away to tourists (as that would defeat the entire purpose of price discrimination). Let's assume that Hamilton Gardens has a good way of doing these things (and we'll come back to that later).

Since those conditions are met, Hamilton Gardens can price discriminate. This is shown in the diagrams below. Both diagrams show a firm with market power (Hamilton Gardens), and each diagram corresponds to one of the sub-markets. The sub-market on the left represents the locals, who have more elastic demand - notice that the demand curve D1 is relatively flat (which means that a change in price will have a big effect on the quantity that these consumers demand). The sub-market on the right represents the tourists, who have less elastic demand - notice that the demand curve D2 is relatively steep (which means that the same change in price would have a smaller effect on the quantity that these consumers demand, than it would for the locals). The marginal cost (MC) is the same in both sub-markets - as noted earlier, it doesn't cost Hamilton Gardens any more to provide entry to a local than it does for a tourist. [*]

Hamilton Gardens will maximise profits by selling the quantity where marginal revenue (MR) is equal to marginal cost (MC) - this is the standard short-run profit-maximising condition. In the tourists sub-market, the profit-maximising quantity occurs where MR2=MC, which is Q2. In order to sell that quantity in the tourists sub-market, Hamilton Gardens should set the price equal to P2. The problem with that high price P2 is that in the locals sub-market, no consumers would be willing to visit Hamilton Gardens at all. Hamilton Gardens can increase profits if it charges a different price in the locals sub-market, from the price it charges in the tourists sub-market. In the locals sub-market, the profit-maximising quantity occurs where MR1=MC, which is Q1. To sell that quantity in the locals sub-market, Hamilton Gardens should set the price equal to P1. 

So, profit maximising across both of these sub-markets would require Hamilton Gardens to sell to the locals sub-market at a low price (P1), which may be equal to zero, while at the same time selling the same entry to the tourists sub-market at a high price (P2).

Now, there are some problems with this approach. Hamilton Gardens needs to be able to tell Hamilton residents apart from tourists. There are limited practical ways to achieve this. Perhaps Hamilton City Council issues an ID card to every resident? That would be effective, but expensive. So, perhaps Hamilton Gardens simply asks residents to bring proof of residency with them to get free entry. That could be a rates bill, or an electricity bill, or similar, with their residential address on it. Hamilton Gardens could, in theory, then match the name on the bill with some other form of ID, to make sure that the person wanting free entry is the same person named on the proof of address. I doubt they would do this, because it will take a lot of time and effort, and because sometimes you live at an address, but the bills are not in your name.

So, it's likely that all that will be required is a utility or rates bill with a Hamilton address on it, to get free entry to Hamilton Gardens. That will mean that this attempt at price discrimination doesn't strictly meet the second condition for price discrimination - Hamilton Gardens won't effectively know who is a resident, because any Hamilton resident can give a proof of address to a tourist, which would allow the tourist free entry to Hamilton Gardens. None of that means that price discrimination will fail entirely here. Movie theatres don't check ID before allowing entry, and they seem to get away with price discrimination just fine [**].

*****

[*] Notice that we are drawing a constant-cost firm here (so marginal cost is equal to average cost, and all units cost the same to produce and sell). That makes our explanations a little easier than the case where marginal cost is increasing.

[**] As far as I know, no student has ever taken up the business opportunity I point out every year, to buy tickets for people who would otherwise pay general admission at movie theatres, and pocket (some of) the difference in prices.

Read more:

• Are tourism operators in NZ missing a trick?
• Price discrimination in tourism... India edition
• Price discrimination and the Great Walks

ChatGPT as an academic author

ChatGPT and subsequent large language models are going to change the world of writing. Having a tool that reduces some of the drudgery of drafting writing will be a big help for people who do a lot of writing. Academics do a lot of writing, so perhaps there are opportunities for ChatGPT to contribute to academic writing. I've played around a little, and I can see that these tools would be useful for writing quick outlines that can be re-written later. The extent of hallucination and made-up facts makes ChatGPT infeasible to fully replace the traditional task of writing for now. The good and the bad of ChatGPT as an academic author were highlighted earlier this week, when Gail Pacheco gave us an example of a spoof economics paper written by ChatGPT, on the Asymmetric Information blog.

I suggest reading Gail's ChatGPT paper. The prompt she used was "Can you write a fake economics paper on ‘The efficiency of using AI to write an economic paper’, including a table and references". ChatGPT fulfilled the requirements to the letter. However, the statistics were entirely made up, as were the references (and Gail even checked them!). Putting aside the hallucinated facts, the writing itself was fairly good, and possibly indistinguishable in terms of quality from some papers that I have reviewed, written by PhD students and/or authors whose first language is not English. This suggests that, as I suspected, ChatGPT might be a good tool for creating a first draft, or a broad outline, of a paper.

How much can ChatGPT help? For now, not much. It can't draw on the real academic literature, so literature review and introductory sections will require quite a bit of work. However, in future that might be addressed if large language models have access to Google Scholar or web searches.

Also, the made-up facts and statistics are problematic. However, if future large language models can draw from actual statistics generated by the human author using statistical software, then suddenly the task of writing papers changes dramatically. Imagine exporting regression tables from Stata or R, handing them over to ChatGPT, and having ChatGPT write a serviceable first draft of the results section of a paper, or a discussion of the results linking back to the previous literature cited in the earlier sections of the paper. That would be a game changer.

Journal editors (of which I am one) had better be ready for a flood of AI-generated papers (with real analysis included, or not!). It's already happening in fiction publishing. Academic publishing will not be too far behind. For now, the quality of those papers will be pretty bad, and hopefully they won't pass the desk-reject stage. If they do, competent reviewers should be able to recognise any made-up statistics or references, and reject the paper. That will not be quite so simple in the future.

Artificial intelligence is going to not only change teaching and assessment, but it is going to change the way that we communicate our research. The robots are coming for our (academic) jobs!

Economic problems generated in the wake of disaster

At the end of last year, I quipped that 2022 had felt like the year of the shortage. It hasn't finished yet. In the wake of Cyclone Gabrielle, many parts of the North Island lost electricity service, increasing the demand for generators. As the New Zealand Herald reported this morning:

The cyclone has also seen Northlanders without power rushing to stores and buying generators to power their homes.

Donovans Trade Supplies manager Scotty Brown said his business sold 200 generators in four days, which was more than they would usually sell in a year.

“One person looked after only generators, and it was mental, really mental. And the person was me,” Brown said.

On Monday last week when the storm was in its early stages, Brown sent a truck and a driver down to a supply branch in Auckland to pick up 60 generators,

“We unloaded them out the back and by lunchtime Tuesday, they were all gone,” Brown said.

On Wednesday, the team went to Auckland again to pick up 120 generators and “within a day and a bit” they had sold them in Whangārei.

When demand increases, we would expect prices to increase, as shown in the diagram below. When demand increases from D0 to D1, the equilibrium price increases from P0 to P1, and the quantity of generators traded increases from Q0 to Q1.

Indeed, that appears to have happened in some cases:

Christchurch company All Trade also flew 50 to 60 generators up to Donovans Trade for Northlanders to purchase when the other supplier ran out.

Brown said customers were thankful to receive generators and get power to their homes back, and they have only 20 generators left in stock.

He said unlike Donovans Trade, he’d heard some places had been increasing the prices of generators due to the demand.

“We didn’t want to take advantage of the situation.”

That last point, that All Trade didn't want to take advantage of the situation, is surprisingly common among sellers. It is inconsistent with what we would expect from the demand and supply model shown above, but it is consistent with how consumers perceive price changes.

In a famous study (ungated), Nobel Prize winner Daniel Kahneman (along with fellow Nobel Prize winner Richard Thaler, and Jack Knetsch) found that when a hypothetical hardware store raised prices for snow shovels following a snowstorm, consumers felt strongly that the price increase was unfair. In contrast, it was perceived as fair for stores to raise prices when their costs increased. Raising prices to increase profits in the short run, taking advantage of market conditions like a snowstorm (or a cyclone), may actually be inconsistent with long run profitability. If a firm develops a reputation for price gouging, the negative image among consumers may actually harm the firm overall.

But keeping prices low creates a different problem - a shortage. In the diagram above, if firms kept the price at the original level of P0, the quantity of generators demanded would be QD, but there would still be only Q0 generators supplied. The difference between QD and Q0 is the shortage. More buyers want generators at the low price P0 than there are generators available. Some willing buyers will miss out on a generator.

The Northland sellers have solved this problem partly by increasing the supply of generators (by bringing in stocks of generators from elsewhere in the country). However, if the shortage is large enough, even that approach would not entirely solve the problem. Most likely the limited number of generators will go first to whoever is quickest to call a supplier, or to whoever knows a supplier personally. Everyone else will miss out.

So, we face an economic problem when disaster strikes. Either prices rise in the wake of the increased demand for generators, or there is a shortage of generators. It might be tempting to think that the government can provide a solution here that keeps prices low (for example, by passing laws against price gouging), while imposing a better way of allocating the limited number of generators. Maybe, instead of going to whoever knows a supplier personally, the government could allocate the generators to those most 'in need'. But, such a subjective approach is open to all sorts of abuse. That is why most economists favour letting the market solve the shortage, by allowing prices to rise. Generators would then be allocated to those that need them the most, without anyone having to decide who has the greatest need. That's because those with the greatest need should be willing to pay the most for a generator. So, when prices increase, the buyers who have the least need for a generator, will be the first to drop out of the market, leaving the generators for those with the greatest need.

There is one final problem here, which is that willingness-to-pay for a generator is not only determined by need. It is also determined by income, because those with highest incomes generally tend to be willing to pay more for goods. The final distribution of generators determined by the market would tend to result in generators going to higher income recipients in a way that many people would consider an unfair allocation. At least, that is if generators were being bought by households. The highest income buyers of generators, with the highest willingness-to-pay, are actually likely to be businesses, trying to keep the lights on and serve their customers. An allocation of generators, that sees them going to supermarkets and service stations first, is probably what most people would prefer.

So, there is no perfect solution to the problem that the cyclone generated (pun intended!). Prices can be kept low and there will not be enough generators for everyone who is willing to buy one. Key service businesses may miss out on generators that are snapped up by buyers who have a close relationship with the sellers. Alternatively, the market can allocate generators, which would result in higher prices but no shortage of generators (at least, at the higher market price that would result). And, it is more likely that key service businesses would end up with those generators. All without the government needing to intervene. You can probably see now why economists might prefer price gouging as a solution.

The health effects of Swedish prisons

Does time in prison make people healthier, or less healthy? On the one hand, prisons are a challenging environment. They are stressful, and the risks of harm through violence are high. As we've discovered recently, they are also an excellent super-spreading environment for infectious diseases. And prisoners may suffer from reduced nutrition, or reduced access to health care. On the other hand, perhaps prisoners' nutrition and/or access to health care may be improved by being incarcerated. Many prisoners have untreated (or undiagnosed) mental health problems, or substance abuse problems, that can be more effectively treated in an institutional setting.

So, which is it? No doubt, it depends on the particular prison context, and the health of the prison population at the time they go into prison. Let's take a particular prison context: Sweden, which has an excellent reputation for rehabilitation (see here and here), and a relatively low prison population (74 per 100,000 population, compared with 155 in New Zealand, and 505 in the US). This recent article by Randi Hjalmarsson (University of Gothenburg) and Matthew Lindquist (Stockholm University), published in the American Economic Journal: Applied Economics (ungated earlier version here), looks at the impact of time in Swedish prisons on health outcomes.

Hjalmarsson and Lindquist make use of a neat natural experiment, being:

...Sweden’s 1993 and 1999 early release reforms, which held sentences constant but increased the share of time inmates were required to serve from 50 percent to 67 percent. Exposure to the two- thirds reform depended on the date of conviction and sentence length. Shorter sentences (4–12 months) were fully treated by the first reform and longer sentences (≥ 24 months) by the latter; intermediate sentences were partially treated by both.

Because time in prison changed, but prison sentences did not, Hjalmarsson and Lindquist look at how the increase in the number of days in prison affects health, while holding sentence length (and therefore, the severity of crime the prisoner is being sentenced for) constant. Their final sample consists of nearly 47,000 sentences of between 4 and 48 months, which commenced between 1992 and 2001. Because Swedes have an effective population register, Hjalmarsson and Lindquist are able to link prisoner records with hospital and mortality data. Looking at the impacts up to ten years after release from prison, they find a variety of impacts, including that:

...exposure to the two-thirds reform does not harm post-release health and actually improves it. Though the reduction in mortality risk is not quite significant when looking at the entire sample, these aggregate results mask important heterogeneity in two dimensions. First, significant reductions in the overall chance of death (especially in the first two years post release) are seen for positively selected subsamples, including those with no past prison exposure, property offenders, relatively young offenders, and those with some past employment. Second, significant effects are seen for the whole sample when zooming in on causes of death particularly relevant for this population. There is a large, significant, and immediate reduction in the chance of suicide; the chance of suicide is still reduced by 38 percent ten years after release. These suicide results are driven by individuals with previously identified mental health issues and by violent offenders.

Taken all together, these are positive results for the Swedish prison system. Why does it do so well? Hjalmarsson and Lindquist can't definitively tell, but note that:

First, health care in Swedish prisons is of high quality. Second, more time in prison is positively related to visits with medical professionals (doctors, nurses, and psychologists), medication, and starting and completing treatment programs. High-quality health care and treatment that increases with time served is consistent with our findings of the health-improving effects of the reform.

So, if we want to improve the health of prisoners, should we be keeping them in prison longer? That would probably be extending these results too far. Remember that context matters. As Hjalmarsson and Lindquist note, the health care available in the Swedish prisons is high quality, and prisoners access it readily. That is not the case in all prison settings. So, we shouldn't use these results to conclude that prisons improve health, but rather that Swedish prisons improve health, and that moving towards the Swedish model may have positive impacts. Of course, then we run into other problems, because Hjalmarsson and Lindquist also note that:

...Sweden spends more money per inmate than any other country...

If we want better health incomes for prisoners, this comes with an increase in cost. We'd need to weigh up those costs and benefits to make a sensible decision about what is best to do. As the saying goes, there is no such thing as a free lunch (as my ECONS101 students will learn when teaching starts next week! [*]).

[HT: Marginal Revolution, last year]

*****

[*] Not literally though. I'm not offering them free food. Instead, we will cover the concept of opportunity cost in the second lecture.

Local restrictions against unpopular retailers

When the Sale and Supply of Alcohol Act 2012 came into force, it gave local councils the ability to develop local alcohol policies (LAPs), to control the availability of alcohol in their city or district. One of the things that LAPs could do was to restrict alcohol licenses from being granted for premises that were in close proximity to 'sensitive sites' (like schools, churches, alcohol treatment providers, etc.). At the time, there was a lot of talk from some advocates about creating wide exclusion zones around these sensitive sites. I pointed out (to a number of people) that these exclusion zones wouldn't have to be very big in order to create a de facto ban on alcohol licenses entirely. Sanity prevailed, and those councils that have these sorts of restrictions in their LAPs have not made them excessively large.

Fast-forward to 2023, and the same arguments for exclusion zones are being made in relation to vape stores. For example, the Asthma and Respiratory Foundation was last year calling for a ban on retailers selling vaping products within one kilometre of a school. How feasible is that sort of control?

Steve at City Beautiful did the GIS work and reported it last year. Here's their map of all of the areas where a vape shop could set up, if you exclude all areas that are within one kilometre of a school, and exclude all areas that are not zoned commercial:

The grey areas are places where a vape shop could not be placed. The orange areas are the few places where a vape shop could be placed, in a commercial zone and more than one kilometre from a school. That's right. Almost nowhere could have a vape shop. As Steve notes in his post:

Yes, only those few tiny scraps of land shown in orange are where vape shops could go. A couple of remote industrial or business areas, sometimes with only a single site available. If there’s already a different shop there? Tough luck. The only significant area where a vape shop could actually go is right at the northern end of the city centre - and rather than being a realistic idea for every single vape user in central Auckland to come downtown every time they need to stock up, it just shows more than anything else, how desperately Auckland’s city centre lacks a school!

Urban planning is hard. Regulations that seem sensible can have unintended (as well as intended!) consequences. The problem here is that commercial zones tend to be right next to schools. If you think about your neighbourhood shops, I bet that there is a school next door to them, or just around the corner from them. Restricting unpopular retailers from being near to schools is essentially the same as restricting them from operating in most of the commercial zones in the town or city. And proponents of these regulations forget the smallest towns, of the kind that have a single school, and a single block of shops, usually on the same small stretch of main road.

There are better ways to control demerit goods than to effectively ban them. License the sellers and restrict the number of licences, tax the sale of the product, impose minimum prices, have stringent age restrictions, or do some combination of all of these things. A de facto ban, masquerading as an exclusion zone that only applies around schools, is not the way.

And besides, the location of stores selling these products becomes almost irrelevant when you consider online sales and deliveries. These products can be purchased anywhere, with very little in the way of controls. The last couple of weeks I've been conducting fieldwork looking at same-day delivery of alcohol (and that's why this blog has been pretty quiet of late). It's not quite the Wild West, but it's far from ideal at the moment. More on that in a future post.

[HT: Eric Crampton at Offsetting Behaviour]

Rainfall, military food shortages and the assassination of Roman emperors

Dictators maintain control of their country by maintaining control of the military. If they lose control of the military, they often lose control of their country, or worse. This 2018 article by Cornelius Christian (Brock University) and Liam Elbourne (St. Francis Xavier University), published in the journal Economics Letters (sorry, I don't see an ungated version online), looks at cases where Roman emperors lost control, and their lives (through assassination).

Christian and Elbourne first note some key facts drawn from past research:

(1) The Roman economy was largely agricultural, depending on rainfed agriculture (Harper, 2017).

(2) The bulk of the Roman army was stationed along the Western frontier, and relied heavily on local food sources (Roth, 1998; Elton, 1996)...

(3) Food transport, in Ancient Rome, was very slow (Terpstra, 2013).

Christian and Elbourne then infer that shocks to military food supply, arising from low rainfall in the frontier areas like Germania, would reduce support for the emperor, increasing the chances that the emperor is assassinated. They use data covering the period from 27 BCE to 476 CE, which covers the period from the beginning of the reign of Augustus to the fall of the Western Roman Empire. Over that period, the relationship between rainfall and the number of assassinations is captured in Figure 1 from the article:

If you squint really hard, you can probably see that there are more assassinations in times when rainfall is lower. This is most apparent in the middle of the study period, between 200 and 300 CE, when there was a century of relatively low rainfall and many assassinations. Turning to their quantitative results, Christian and Elbourne find that:

...a standard deviation decline in rainfall causes an 11.6% standard deviation increase in assassination probability... [and] a standard deviation drop in rainfall causes a 13.4% standard deviation increase in total assassinations.

Now, of course this paper doesn't definitively establish a causal relationship between rainfall and assassinations. However, it is interesting that when, instead of using rainfall from the year before, Christian and Elbourne use rainfall from the following year, the correlation with assassinations becomes statistically insignificant. They also find that rainfall is a statistically significant predictor of frontier mutinies, which is consistent with rainfall increasing disquiet in the military.

However, the parts of the story included in this article don't quite feel like they add up to a coherent whole to me. The correlations are strong, but not necessarily entirely convincing. If the mechanism through which negative rainfall shocks increase emperor assassinations is through declining military support, why don't Christian and Elbourne use an instrumental variables approach, with rainfall as an instrument for frontier mutinies, and frontier mutinies as an endogenous variable explaining assassinations? Surely they tried this approach, so there being no mention of it in the paper is perhaps a signal to us that the approach doesn't work. In that case, we should be a little sceptical of the results.

Nevertheless, it is an interesting paper, and I'm sure there will be more work in this area. As attributed to both Napoleon Bonaparte and Frederick the Great, "an army marches on its stomach". When the stomach is not full, I guess the army stops marching, and eventually turns on its leaders.

Big data and artificial intelligence may not make pure command or socialist economies any more feasible

Economic systems can be considered to fall along a continuum. At one end is the pure market economy, where the government only provides the basic rule of law, enforcement of contracts and property rights, and some public goods. All property and the means of production are private owned. At the other end is the pure command economy, where the government is responsible for all key production and consumption decisions, and owns most property and the means of production, on behalf of citizens.

All modern economies fall somewhere in between these two extremes, and most developed countries are closer to the market economy end of the continuum than the command economy end of the continuum. That shouldn't be a surprise. Market economies have been incredibly successful over the past century and more, while socialist and command economies have mostly fallen away. [*] However, long before it was apparent that the command economy was a slow-motion train wreck, it was heavily critiqued by the Austrian school economist Ludwig von Mises, in his famous paper Economic Calculation in the Socialist Commonwealth (originally published in 1919, then translated to English in 1920). Von Mises essentially argued that money and money prices provided the information that was necessary for the market system to function efficiently. Because a socialist economy lacked money prices, a socialist planner could not properly coordinate the economy in as efficient a manner as would be achieved in the market economy. Von Mises concluded that "rational economic activity is impossible in a socialist commonwealth".

The difficulty of rational economic planning by a central planner was brought home later by economics Nobel Prize winner Leonid Kantorovich. In a story I first read in Tim Harford's book Adapt (which I reviewed here), Kantorovich applied linear programming methods to the production and allocation of Soviet steel for a single year (it is mentioned briefly in this biography of Kantorovich as well). Solving this problem using complex mathematics took several years. And that was for a single (albeit large) industry. For a single year.

However, computing power and computer science methods have progressed significantly since von Mises' and Kantorovich's time. The availability of big data, machine learning algorithms, and artificial intelligence might suggest that command economy planning will become more feasible. Not so, according to this new article by Karras Lambert (George Mason University) and Tate Fegley (Montreat College), published in the Journal of Economic Behavior and Organization (sorry, I don't see an ungated version online).

Lambert and Fegley go back to the original arguments of von Mises and outline how the basic problem of economic calculation is not resolved by big data, increased computation, or artificial intelligence. Specifically, they note that:

Can “big data”, absent the market pricing process, generate a commensurate cardinal unit with which to meaningfully compare costs and proceeds of productive actions? Does it provide a mechanism for translating ordinal preferences into commensurate cardinal numbers? Does it give a central planner the capability to engage in entrepreneurial price appraisal without genuine market prices for goods of all orders? The answers are evidently no on all counts. As a result, “big data” itself, while undoubtedly a useful tool for entrepreneurs operating within a market system, cannot replace the pricing process or render unnecessary private ownership and ex- change of the means of production.

The key point for me is that big data and algorithms, by construction, are based on past data. The most useful data for planning are data that come from the decisions of consumers and producers. If we moved from a market economy to a command economy, the extant data would become progressively outdated over time. Eventually, relevant data necessary to estimate big data models and algorithms would not exist at all.

It seems to me the argument about whether the pure socialist or command economy is feasible needs to be set aside for a while yet. Perhaps we come back to it when we get artificial general intelligence, or where we get so much computing power that we can simulate entire people, with all of their preferences, quirks, and sub-optimal decision-making capacities.

*****

[*] Some might argue that China is a counter-example. However, when you look at the characteristics of the Chinese economic system, it looks much more like a market economy than a command economy. Indeed, it is not so much a socialist or command economy, but socialism with Chinese characteristics.

Gender differences in reactions to editorial decisions in economics

There is a persistent gender gap in economics, which starts at undergraduate level and extends through graduate study, junior academic positions, tenure, and to the professoriate. The gender gap gets worse as you move up through each level, with women making up a smaller proportion as you go. Many studies have now looked into this, but one aspect is relatively under-explored - how different genders respond to setbacks in the publishing process.

That research gap is what makes this recent working paper by Gauri Kartini Shastry and Olga Shurchkov (both Wellesley College) most interesting. They presented a sample of around 1300 academic economists with a hypothetical scenario regarding the rejection of a paper. Specifically:

Respondents then read a hypothetical decision letter from an editor which begins by describing referee reactions to a paper the respondent hypothetically submitted for publication at a top general-interest journal. Our experimental manipulation randomizes respondents into three main treatments: a revise and resubmit decision (R&R), a reject and resubmit decision (RJR), and a flat rejection (FR); this randomization affects only the last few lines of the decision letter. A key feature of the design is that outcomes are measured following almost identical decision letters from an editor on a hypothetical paper, ensuring that all respondents are given the same information about the quality of this submission. We are interested in the differential effect of the more negative treatments (FR/RJR) relative to the baseline (weak R&R) on the respondents’ perceived likelihood of eventually publishing the paper in a highly-regarded journal.

Shastry and Shurchkov find that:

...getting an RJR reduces the perceived likelihood of publishing the paper in any leading journal relative to an R&R, but getting a rejection has the most negative impact. On average, negative decisions have similar effects on men and women. However, we observe heterogeneous effects by rank: female assistant professors who get a rejection perceive an 18 pp lower likelihood of publishing the paper in any leading journal than male assistant professors, as compared to the difference across genders for those who get an R&R. The gender gap is not present among tenured associate and full professors.

Shastry and Shurchkov try to tease out the mechanism explaining these results, positing that:

...female assistant professors attribute the negative feedback of a rejection to subpar quality of their work to a greater extent than men do, and that this is exacerbated by the time constraint of an upcoming review.

There is some evidence to support this explanation. However, the more interesting (to me) finding is that the gender gap closes after tenure. On this point, I strongly suspect that there is some survivorship bias. That is, economists who are less resilient to negative feedback may be less likely to achieve tenure, and if male economists are (because they are more confident, or some other reason) more resilient, then more male economists would survive to tenure than female economists. And both male and female economists who survive to tenure will tend to be similar in terms of how they respond to negative feedback. That's speculative on my part. Shastry and Shurchkov have no evidence in favour of it, but neither are they able to discount it as a possibility.

This paper is useful in filling in a bit more detail on how the gender gap in economics arises. However, it would be useful to build on this work in order to better understand the mechanisms that underlie it, because it is the mechanisms that are the source of the problem.

[HT: Marginal Revolution, last year]

What machine learning is telling us about the gender dynamics in economics seminars

Economics seminars have come in for a lot of criticism for apparent gender bias (see here, or for a broader view on gender in economics see the links at the end of this post). To some extent, past research on gender dynamics in economics seminars has been limited by the hand-coding of data, so limited detail on the interactions within the seminar are available to analyse. Not any more, thanks to machine learning, as explained in two new papers.

The first paper, by Amy Handlan and Haoyu Sheng (both Brown University), uses machine learning for audio analysis of presentations at the 2022 NBER Summer Institute. The audio classification algorithm that they use allows them to automatically classify speakers by gender, age, and the tone of their comments. They find that:

...women in the audience are more likely to ask female presenters questions than male presenters, and female presenters are more likely to be assigned female discussants.

This suggests that there is significant gender sorting (although Handlan and Sheng can't say anything about the wider audience in each presentation, because they only have data on audience members who spoke). Interestingly, next they find that:

Regarding interruptions, we find that there are similar number of interruptions for both male and female presenters.

That is somewhat at odds with earlier research (see here, for example). However, more consistent with the earlier research, Handlan and Sheng also find that:

...interruptions for female presenters last longer than those for male presenters...

But what about tone? In that respect, they find:

...gender differences in tone within speakers. On average, female speakers are more likely to sound positive or happy, while male speakers are more likely to sound negative and serious or angry. This holds whether we consider only presenters or only speaking audience members. Furthermore, when we consider how speakers may change their tone over time, we find that tone is highly persistent for both men and women. That is, if you sound happy and are uninterrupted, you are likely to continue sounding happy.

So far, so good. But what happens when speakers are interrupted, or responding to others? In that case:

...we find that speakers sound more negative when responding to women, whether the person is an audience member asking a question to a female speaker or a presenter responding to a female audience member. When we look at the interaction between presenter gender and audience gender, we find that female speakers respond more negatively to other women compared to how they respond to men. The gap in tone responding to men versus women is larger for female speakers than for male speakers.

This finding that women are harsher towards other women than men are towards women, is unfortunately looking increasingly common in economics. Handlan and Sheng suggest that this may be due to societal norms that lead to higher expectations of women, as well as because:

Women may speak more positively to men in an attempt to offset a larger negative bias from men compared to women.

I'd label those reasons speculative for now. Handlan and Sheng aren't able to test them, of course. However, the paper has some interesting insights, and not just about gender. For instance, people (presenters and audience members) in macroeconomics seminars are more likely to sound sad, but less likely to sound angry, than people in microeconomics seminars. Given that this was 2022, historically high inflation might make macroeconomists sad, and microeconomists angry? We need some causal analysis of tone!

Anyway, moving on. The second paper I want to discuss is this job market paper by Mateo Seré (University of Antwerp). His dataset is much broader than Handlan and Sheng's, covering all web-streamed (on YouTube) seminars that were part of a seminar series of a university economics department in the US or Europe, over the period from 2020 to 2022 (as well as the National Bureau of Economic Research (NBER), the American Economic Association (AEA), and the Centre for Economic Policy Research (CEPR)). Seré focuses specifically on interruptions to the seminar, and first notes that:

...the distribution of interruptions made during seminars presented by women is slightly shifted to the right compared to the distribution for males.

Specifically, female presenters receive between 0.9 and 1.5 more interruptions than male presenters, controlling for presenter characteristics and seminar characteristics. That is a large effect, given that the mean number of interruptions is 11. So, female presenters receive about 10 percent more interruptions than male presenters. That is inconsistent with Handlan and Sheng's results, but remember that their results come from the NBER Summer Institute only, whereas Seré is looking at a much broader set of seminar series.

When looking at who does the interrupting, Seré finds that:

...the proportion of female interruptions is significant to explain the overall number of interruptions in a seminar only when the presenter is a woman.

In other words, the additional interruptions that female speakers experience are mostly driven by female audience members. Seré notes that male audience members ask female presenters 0.2 more questions than they ask male presenters, but female audience members ask female presenters 1.0 more questions than they ask male presenters. Given than most audience members are male (and so they ask more questions overall), this is a substantial difference.

Seré then finds that there is a difference in the nature of interruptions between male and female audience members:

Being a female presenter is related with an increase in the number of questions from females in the audience and with a decrease in the ones made by males. Furthermore, while men in the audience make more comments on average when the presenter is female, the gender of the presenter has no effect on the number of comments made by women.

This is interesting, as it is closest to the results on tone from Handlan and Sheng. Questions require a response, whereas comments do not. Looking across both studies, if female audience members ask more questions of female presenters rather than making comments, perhaps those questions come across as more negative in tone?

Together, these two papers paint an interesting picture of some of the dynamics within economics seminars. Clearly, this is just the beginning of research in this area. In particular, Handlan and Sheng have made their algorithm available for others to use for coding and analysing other seminar series. Expect more research in this space in the future.

[HT: Marginal Revolution for the Handlan and Sheng paper, and Development Impact for the Seré paper]

Read more:

• The gender gap in economics
• More on the gender gap in economics
• CORE on 'missing women in economics'
• Female representation in economics
• Justin Wolfers on why we should care about the gender gap in economics
• Why we should care about the gender gap in economics
• Could closing the gender gap in economics be as simple as providing students with information?
• A sobering report on the culture in the economics profession
• The gender gap in reviewing and editing for top economics journals
• The gender gap in U.S. economics education
• UWE and the gender gap in economics (results pending?)
• Three papers on the gender gap in economics seminars
• Disappointing results on simple information interventions to close the gender gap in economics
• Gender differences in fields of specialisation within economics
• Gendered citations of papers in the 'top five' economics journals
• The positive effect of mentoring for female economists

Mendelian randomisation doesn't necessarily overturn the alcohol J-curve

The alcohol J-curve is the common empirical finding that moderate drinkers have better health than abstainers, and better health than heavy drinkers (see this post, for example). If you plot the relationship between the amount a person drinks and negative measures of health, the resulting curve is shaped like the letter J (as in that earlier post). However, few of the studies that establish a J-curve relationship demonstrate a causal relationship. That's because it is difficult to randomise people into a level of drinking.

However, a relatively new development in epidemiology is the idea of Mendelian randomisation. People are randomly assigned genes at birth. Some of those genes are associated with alcohol consumption. So, alcohol consumption is (partially) randomly assigned by the assignment of genes. Studies can then use instrumental variables regression (a relatively common technique in economics) to estimate the causal effect of alcohol consumption on a range of health (and other) outcomes.

What happens to the J-curve in these Mendelian randomisation studies? This editorial published in the journal Addiction in 2015 (open access), by Tanya Chikritzhs (Curtin University) and co-authors, summarises the state of knowledge up to that point. They note that there is no J-curve relationship observed in Mendelian randomisation studies, or at least that the results are much more equivocal about its existence, and conclude that:

The foundations of the hypothesis for protective effects of low-dose alcohol have now been so undermined that in our opinion the field is due for a major repositioning of the status of moderate alcohol consumption as protective.

I was recently referred to this editorial during a discussion on the J-curve. Having read a bit more about Mendelian randomisation though, I am not entirely convinced. The problem is that in these Mendelian randomisation studies, the two main assumptions of instrumental variables analysis must be met. First, the instrument (having the gene, or not) must be associated with the endogenous variable (alcohol consumption). That assumption should be relatively easy to meet. A researcher simply searches the literature on genome-wide association studies for some gene that is associated with alcohol consumption. Second, the instrument must only affect the dependent variable (health outcomes) through its effect on the endogenous variable (alcohol consumption), and not directly or through any other variable. This is known in economics as the exclusion restriction.

The exclusion restriction is a difficult to satisfy, in part because it is impossible to test statistically. Instead, most researchers settle for being able to identify an instrument that is 'plausibly exogenous'. That is, they find an instrument that is extremely unlikely to affect the outcome variable directly, or indirectly through any other mechanism than through the exogenous variable. In this case, that would mean identifying a gene that could only possibly affect health outcomes through its effect on alcohol consumption.

And that is the problem here. There is no gene for alcohol consumption. All that genome-wide association studies will identify is genes that are associated with alcohol consumption. Those genes all have some other purpose, and that other purpose may be linked to health outcomes in a way that doesn't involve alcohol consumption. As far as I am aware, the Mendelian randomisation studies to date haven't been able to establish that the gene in question has no other effects on health outcomes. That makes their claims to causality no better than those of the correlational studies that they are supposed to improve on.

Mendelian randomisation is good in theory, but not always in practice. For now, the J-curve lives.

Academic achievement and class scheduling

It's only a few weeks until classes start for 2023. For the first time in many years, I will have a lecture scheduled in the dreaded Monday 9am timeslot. Will any students show up so early in the week? And if they do, will they be so bleary-eyed from a weekend of working and partying that it negatively impacts their learning?

I'll find out an answer to the first question on the first day of classes. Given my experience last trimester, I'm not holding out hope for high attendance (although that will seriously be to the detriment of the students - more on that in a future post, as I am currently looking into how students' engagement in my classes affected their performance). As for the second question, this 2018 article by Kevin Williams (Occidental College) and Teny Shapiro (Slack, Research & Analytics), published in the journal Economics of Education Review (sorry I don't see an ungated version online), provides some answer.

Williams and Shapiro use data from the US Air Force Academy, where students:

...alternate daily between two class schedules within the same semester. Students have a similar academic course load, but the alternating schedule creates variation in how much time students spend in class on a given schedule-day. This allows us to assess how a student performs with one schedule relative to their own performance with a different schedule.

Also, since students are randomly assigned to instructors and schedules, this provides an opportunity to estimate the causal effect of different classroom schedules on students' academic performance. This random assignment has made the USAFA data a popular choice in the economics of education (see here, for example). In terms of scheduling:

USAFA runs on an M/T schedule. On M days, students have one set of classes and on T days they have a different set of classes. The M/T schedules alternate days of the week...

Williams and Shapiro observe that there are two effects of class scheduling on student academic performance:

The first is the cognitive load a student has experienced before the start of a class. We refer to this as the student fatigue effect or cognitive fatigue. The second is the timing of a class: students may perform less well if classes are scheduled when they’re naturally less alert. We refer to this as the time-of-day effect. We expect student fatigue to unambiguously hinder academic performance. The time-of-day effect may vary throughout the day.

Williams and Shapiro use data from 6981 students (of whom 4788 are freshmen), and in total have over 230,000 course-level observations (of which over 180,000 are core courses). The key outcome is students' normalised grade, while the explanatory variables that Williams and Shapiro are most interested in are the number of consecutive classes and the number of cumulative classes on that day (to measure the student fatigue effect), and the time of day (to measure the time-of-day effect). They find that:

Consecutive classes have a consistently negative impact on performance... A student sitting in their second consecutive class is expected to perform 0.031 standard deviations worse than if she took the same course after a break. We take this as solid evidence of cognitive fatigue. When student’s schedules require them to sit in multiple classes in a row, they perform significantly worse in the latter classes, likely because of a decreased ability to absorb material.

The effect of cumulative classes, the total number of prior courses a student has taken on a given day, varies more across our models, but is significantly negative in our preferred specification... This suggests that students suffer both from the immediate effect of consecutive classes and the cumulative effect of heavy course loads in a single day...

The penalty for students taking a 7 am hour is consistent and robust. All time-of-day coefficients are positive and significant, which suggests holding class every hour of the day after the first benefits students. These effects are large in magnitude. Students taking a 9 am or later are expected to perform 0.16 standard deviations better than students taking the same class at 7am.

So, there is evidence for both negative student fatigue effects and time-of-day effects, with students performing better when they have classes later in the day. However, given my own predicament, the results are not all bad. Williams and Shapiro compare class times with a 7am class. When you compare with a 9am class, student performance is not substantially worse than later in the day, at least up to 1pm (phew!).

However, not all students are affected the same. When Williams and Shapiro separate students into terciles of academic ability, they find that:

Fatigue has the largest impact on the bottom tercile of USAFA students. A consecutive class reduces a bottom-tercile student’s expected grade by 0.042 standard deviations, compared with 0.030 or 0.019 for top-tercile and middle-tercile students, respectively. Two or more consecutive classes only have a significantly negative impact on bottom tercile students.

As with so many things, the students at the bottom of the ability distribution are the worst affected (although, it is worth noting that the bottom of the USAFA distribution is still students who are in the top 15 percent of all high school graduates).

So, class scheduling does appear to make a difference to student grades. Now, we can't schedule all classes in the late afternoon. But, when there are classes that students actually attend in person (rather than those they mostly watch recorded or online), perhaps those in-person classes could be scheduled later in the day? I might have to make some enquiries before the scheduling of my classes next year.