Sunday, 22 November 2026

This week in research #153

This week I gave a research seminar at the Department of Economics at Cornell University, on the impact of introducing Harriet, our ECONS101 tutor, on student performance and experiences in ECONS101. This was a mildly updated version of talks I gave at the EduLearn Conference and at the University of Exeter earlier this year. The audience was small, but very engaged. There is more to come on this project, as once I have collated the 2026 data, I'll be writing it up for publication.

Aside from that, here's what caught my eye in research over the past week:

  • Süer et al. (open access) find that, when presented with a social dilemma game (like a probabilistic version of the prisoners' dilemma), women's behaviour is more strongly affected by a recommendation on socially optimal behaviour than is men's, which they interpret as women being more 'compliant' than men
  • Brodeur et al. (with ungated earlier version here) map the evolution of economics research from 1998 to 2019, showing that awareness of reproducibility has increased modestly, sample sizes have increased, and econometric methods and causal inference practices have become more sophisticated, but pre-analysis plans, and power tests to accompany null results remain uncommon
  • Harvey (with ungated earlier version here) finds that, in the 1707 Scottish parliament vote for union with England, eligibility for compensation increased the probability of a parliamentarian voting for union by over 30 percent
  • Brülhart et al. (with ungated version here) find that housing rents within 0.7km of an active asylum seeker hosting centre in Switzerland are on average 3.8 percent lower than in the control group, and that the effect is significantly more pronounced when centres host a higher share of asylum seekers from sub-Saharan Africa
  • Barrett and Tan (with ungated earlier version here) find that doubling the rate of immigration at the city level in the US reduces inflation by about 0.1 to 0.2 percentage points, and that the effects are larger for working-age immigrants
  • Dickinson (open access) finds limited evidence that there is a difference in moral choices between daytime (3pm-5pm) and night-time (3am-5am) for choices that harm another participant, but that at night-time there is more dishonesty in a task that doesn't harm other participants

Saturday, 21 November 2026

How much is a four-minute time saving worth?

I've been in the US for the past couple of weeks. I gave a seminar talk at Cornell University this week (more on that in my next post), and my wife and I then drove down to West Virginia for a night, before heading east. Northbound on the I-495 near Bethesda, Maryland, this afternoon, we saw this:

We were stuck in traffic, and that turned out to be a major theme of our travels today. At this time, Google Maps was offering us an alternative route, with a four-minute time saving, but at a toll cost of US$17.52.

A rational driver would only take that option if the value of their time saved (the incremental benefit of the new route) exceeds the additional toll cost (the incremental cost of the new route). So, does a four-minute saving warrant spending $17.52 on tolls? To make the comparison easier, let's convert it to an hourly rate. Four minutes at US$17.52 equates to US$262.80 per hour.

For comparison, someone on a salary of $500,000 per year earns a little less than $262.80 per hour. Maryland's statewide minimum wage is US$15.00 per hour. New Zealand's minimum wage is $23.95 per hour, or about US$13.10 per hour. So, US$262.80 is a pretty steep price to pay. Of course, there were two of us in the car, so we would each only need to value our time at US$131.40 per hour for the toll route to be worthwhile.

Needless to say, professors at New Zealand universities don't earn enough to place that high a value on saving four minutes of travel time. My wife and I took the slightly longer, but much cheaper route. I wonder how many other drivers switched to the more costly route?

Tuesday, 17 November 2026

How the typewriter changed women's work, marriage, and motherhood

In Western countries, one of the overwhelming trends of the 20th Century was the dramatic increase in women's formal labour force participation. The drivers of that change have been both social and economic, but the role of technological change may be underappreciated. A new working paper by Myera Rashid (Northwestern University) focuses on the changes arising from the adoption of the typewriter. While that might not seem like an important technology to drive meaningful change, Rashid notes that:

Offices in the US adapted to the arrival of the commercial typewriter in the 1880s by shifting towards the available supply of educated women who provided cheaper labor: while women initially made up only 12 percent of the secretarial and typing occupation in 1880, by 1900 they accounted for 75 percent of all typists.

Rashid then looks at how the diffusion of the typewriter across counties affects women's labour force participation, marriage rates, and fertility. To estimate causal impacts, she makes use of a shift-share instrument that combines local industry employment shares measured in 1880 (prior to the adoption of the typewriter) with national shifts in the demand for typists between 1880 and 1920. The labour force participation, marriage, and fertility outcomes are all measured in 1920, and the key data she uses come from Census microdata. In terms of labour force participation, Rashid finds that:

...a 1 percentage point (pp) increase in the predicted change in the number of typists between 1880 and 1920 led to a 0.75 pp increase in share of women in the labor force in 1920. Using a back-of-the-envelope calculation, this estimate implies that without the shock of the typewriter, women’s labor force participation rate in 1920 would have been 3.65 pp lower, equivalent to nearly 1.16 million fewer women in the labor force out of a population of 31.7 million women of working age.

Next, Rashid shows that the effects were not just limited to White women, but there were increases in labour force participation among Black women as well. However, how those labour force changes manifested were quite different between White and Black women. Specifically:

...a 1 pp greater increase in the predicted change in typists had a 1.7 pp higher share of Black women in the labor force in 1920. I show that the main mechanism underlying this change is the reallocation of White women out of unpaid housework and into clerical employment, which increased the demand for Black women’s labor in paid household services work.

Turning to impacts on marriage and motherhood, Rashid finds that:

The labor market effects induced by the introduction of the typewriter further led to a 0.4 pp reduction in the share of women who were married and a 0.6 pp decrease in the share of women who had children in 1920. These estimates imply that the share of married women and women with children would have been 2.04 pp and 2.89 pp higher in 1920, equivalent to 650,000 additional married women and 920,000 additional women with children in the absence of the typewriter.

Importantly, these results relate to the proportion of women with children in 1920, rather than their completed fertility (the number of children they have over their lifetimes). The results could therefore reflect women delaying having children, rather than necessarily having fewer children overall. Nevertheless, what Rashid finds are large effects, and she shows that they are robust to various alternative choices or data sources. I particularly appreciated the use of corresponding data from Canada used to construct an alternative shift-share instrument, which alleviates concerns that there are unobserved time-variant changes at the county level that correlate with the shift-share instrument and might have been associated with the outcome variables. 

The next part of the paper is particularly interesting. Rashid extends her analysis to look at social mobility, specifically, she tracks young white women from the 1920 to 1940 Census, looking at how the ranking of their husbands within the income distribution (in 1940) are related to the ranking of their fathers within the income distribution (in 1920), and how that relationship differs for women who are typists from other white women. She finds that:

For women who grew up at the 25th percentile of their father’s income distribution, working as a typist is associated with marrying men with an average income rank that was 18.4 percentiles above the average rank of husbands of non-typists. Importantly, this mobility advantage is not present in alternative white-collar occupations such as teaching, despite the fact that these occupations offered women similar wages.

Using the same shift-share instrument (SSIV) as for the previous analyses, Rashid then finds that:

...conditional on father’s socioeconomic status, women living in counties with a one percentage point greater increase in the predicted change in typists married husbands whose income rank was 1.27 percentiles higher. Among women with fathers at the 25th percentile, residing in a county with an above median value of the SSIV is associated with marrying men with an average income rank that was 4 percentiles higher

Importantly, again these results are robust to alternative methods and data. Importantly, Rashid also compares sisters within the same family, finding that the social mobility advantage of being a typist remains. The mechanism argued to underlie these results is that women working as typists were more able to work alongside men of higher income earning potential than women with a similar education level in other occupations. Rashid specifically contrasts the results with those for teachers, nurses, and telephone operators, finding that none of those other occupations offered the same social mobility advantage than that of typists.

Finally, Rashid draws some parallels in the conclusion between the technological changes brought on by the diffusion of the typewriter and those arising from generative AI. She notes that while there may be displacement effects as a result of new technology, there are also increased labour market opportunities, spillovers between different groups of workers, and social consequences as well. While it's unlikely that generative AI will have a similar effect on marriage and fertility given the baseline conditions are very different now than they were in the late 19th Century, it is difficult to anticipate what, if any, social changes may arise due to the diffusion of this new technology. The history of the typewriter is a useful reminder that the consequences of technological change can extend far beyond the workplace.

Incidentally, Rashid is on the economics job market this year, and according to her website this paper is currently at the second-round revise-and-resubmit stage at the prestigious American Economic Review.

[HT: Marginal Revolution]

Sunday, 15 November 2026

This week in research #152

This week I attended the 73rd North American meetings of the Regional Science Association International (aka the NARSC Conference) in Philadelphia. Philly is a great city to visit, and the conference was excellent. My own presentation was a reprisal of my ERSA Conference paper on the increasing trend towards natural population decrease (more deaths than births) across the territorial authorities and local boards in New Zealand, which will become an increasingly serious policy issue in the future. Here are some of the highlights I found from the conference:

  • Steven Bond-Smith (session 8 in the programme) recognised that working from home has a different geographical distribution from AI exposure, and used that to disentangle the effects of working from home and generative AI on junior jobs, finding that junior jobs have fallen most in larger cities as a result of generative AI
  • Peter Batey (session 35 in the programme) gave an excellent talk about the legacy of Charles Booth, the originator of the social survey, and his maps of London poverty, which represented an application of geodemographics long before geodemographics as a field was even invented
  • John Winters (session 38 in the programme) used data based on women's birth county to overcome the endogenous sorting by location and disentangle fertility trends between metro and non-metro areas in the US, finding that housing costs largely explain the different trends in fertility between metro and non-metro areas
  • Chiara Del Bo (also session 38 in the programme) described the size and distribution of the 'space economy' in Italy, a first step towards developing a satellite account for the space economy
  • Ephraim Nkonya (session 46 in the programme, which was also the session I presented in) showed that US birthrates have declined four times faster in rural than in metro counties over the period from 2019-2023, and that crude birth rates are higher in metro than non-metro areas 
  • Luyi Han (session 59 in the programme) found that broadband internet is positively associated with working from home among non-metro women, but not among non-metro men
Aside from the conference, here's what caught my eye in research over the past week:

  • Chua and Tsiaplias (open access) find that the COVID-19 lockdown in Melbourne had a statistically significant and economically meaningful disinflationary effect on grocery prices
  • Chen et al. find that hosting the Olympics is positively associated with bilateral trade, using a gravity model covering 187 countries from 1950 to 2019
  • Cuesta et al. (open access) develop a theoretical framework for human rights economics

Saturday, 14 November 2026

Try this: Platypus Economics

One of my favourite economists, Justin Wolfers, has started a new venture: Platypus Economics. He says:

Television tells stories but doesn't explain. A ninety-minute lecture explains but doesn't travel. I'd like to build something in between. Work that makes you a little smarter for next time, not just this one.

Actually, it's not really new - it dates back to April. I only just found out about it this week, even though it received coverage in the New York Times in May. Through the site, Wolfers provides links to a YouTube Channel, which has tons of content. This includes series entitled "Diving In", which is a bunch of economics explainers, "Off the Clock", which unpacks the week's economics news, and "The Professor is In", which responds to audience comments and questions. The most recent video (as of yesterday) is entitled "The Most Expensive Shag in the History of Humanity", about the Jensen Huang scandal.

Platypus Economics also has an associated Substack newsletter, and a podcast that is available on all your favourite podcast apps.

Even better, the site has a bunch of Stata worksheets that you can use to replicate the analyses from many of the explainer videos. That makes it an invaluable resource for students (at least, for students who have access to Stata!).

There is so much content to get into. I have barely scratched the surface myself. Enjoy!

[HT: Stata News]

Wednesday, 11 November 2026

How much did slavery cost the American economy?

What was the economic cost of slavery in the United States? Answering that question requires first recognising that the spatial distribution of economic activity could have been very different if there was no slavery. And accounting for that difference is the approach adopted in this 2025 NBER Working Paper (ungated version here) by Treb Allen (Dartmouth College), Winston Chen (Yale University), and Suresh Naidu (Columbia University).

Using microdata from the 1860 US Census, Allen et al. first establish three stylised facts (with several elements to each):

Stylized Fact #1(a): Free workers sort into higher productivity locations, enslaved workers are sorted even more so into higher productivity locations...

Stylized Fact #1(b): Free workers sort into higher amenity locations, enslaved workers are sorted into lower amenity locations...

Stylized Fact #1(c): Free workers sort into locations with less coercion, enslaved workers are sorted into places with more coercion...

Stylized Fact #2(a): There is no correlation between cotton suitability and agricultural employment share for either enslaved or free workers...

Stylized Fact #2(b): Free workers sort out of agriculture in high malaria areas, whereas enslaved workers are sorted into agriculture in high malaria areas...

Stylized Fact #2(c): Both free and enslaved workers are more likely to be employed in agriculture the greater the distance to freedom, but this is especially true for enslaved workers...

Stylized Fact #3(a): Both free workers’ wages and the price of enslaved workers are higher in locations with higher productivities...

Stylized Fact #3(b): free workers’ wages are lower in locations with better amenities, whereas the price of enslaved workers are higher in locations with better amenities...

Stylized Fact #3(c): free workers’ wages are lower in locations with greater coercion, whereas the price of enslaved workers is (slightly) higher in locations with greater coercion.

The first stylised fact shows that enslaved and free workers are geographically sorted differently. While both enslaved and free workers locate more in high-productivity areas, this is for different reasons. Free workers can capture higher wages in those areas, while enslaved workers are more profitable for their owners in those areas. Allen et al. use malaria suitability as a negative indicator of amenity. Free workers avoid areas that are more susceptible to malaria (low amenity), whereas enslaved workers have no choice and cannot avoid those areas. Allen et al. measure the extent of coercion with the distance to the North, since slaves who were located further from the North could be more heavily coerced in work without prompting them to attempt to escape to the North. Unsurprisingly, there are more slaves further from the North, where coercion is easier for slave owners.

The second stylised fact shows that enslaved and free workers are sorted into different sectors of the economy, with enslaved workers more likely to be engaged in agriculture, especially in low amenity (high malaria) areas and areas with more coercion (further from the North). This sorting doesn't simply reflect the suitability of areas for agriculture (specifically, suitability for cotton).

The third stylised fact shows that wages are higher where productivity is higher. Lower wages for free workers in high amenity areas reflects a compensating differential. That is, wages need to be higher in low amenity areas in order to attract free workers. Where coercion is greater (further from the North), free workers earn lower wages as they need to compete with more enslaved workers.

After establishing those stylised facts, Allen et al. turn to a general equilibrium model and use their model to estimate the economic impacts of slavery. In that analysis, they find that:

In equivalent variation terms, slaveowners would have to pay an enslaved worker 1,192% more to get voluntary consent to the allocation of labor induced under slavery.

Breaking that down into three different sources of misallocation, they find that:

...allowing enslaved individuals to choose their own labor supply while remaining in slavery (i.e. removing coercion) yields a welfare gain equivalent to a 361% increase in wages; allowing enslaved workers to retain all of their earnings yields a welfare gain equivalent to a further 81% wage increase; and allowing enslaved workers to choose any occupation in any location in the U.S. (so that they act identically to free workers) leads to a welfare gain equivalent to a 55% wage increase.

Needless to say, those are massive welfare gains from removing slavery, but they likely understate the welfare costs of slavery. Importantly, their model does not capture all of the costs of slavery, including the physical and psychological harms associated with violence, abuse, and the loss of freedom. So, the welfare estimates from the model are best considered as lower-bound estimates of the full welfare costs of slavery (and the gains from removing slavery). It should also be noted that the percentage welfare gains from each source of misallocation compound rather than simply adding together, which explains why the overall welfare gain is so much greater than the sum of the three individual percentages.

Turning to the economy as a whole, Allen et al. then find that:

The movement of formerly enslaved workers out of agriculture causes services and manufacturing sectors to expand: for example, manufacturing output increases by 29% in the North, 19% in the South, and 27% overall. Agriculture expands in the U.S. North by 24%, but contracts in the U.S. South by 29%, for a modest decline of 5.4% overall. All told, aggregate GDP rises by 9.1%, with the reallocation of labor toward higher amenity locations offset by a combination of this structural transformation and the ability of the formerly enslaved to move to the (more productive) U.S. North.

A 9.1 percent increase in GDP is nothing to be sneezed at. Of course, this is based on the structure of the US economy in 1860, and can't be compared to what actually transpired later, after the Civil War.

Interestingly, but not unsurprisingly, not all workers would have been made better off by removing slavery. Free persons' welfare would decline by 0.7 percent overall, reflecting that many previously-enslaved workers would move to high-amenity and high-productivity areas, competing with free workers in those areas and lowering their welfare. However, that welfare loss pales in comparison to the welfare gains that previously-enslaved workers would receive, as well as the overall economic benefits.

US slavery was an immensely profitable institution for southern slave owners. There should be no doubt about that. The wealth accumulated through slavery remains visible today in the large plantation houses dating to the Antebellum period, some of which I may see after I cross the Mason-Dixon Line into the South for a couple of days next week. However, those plantation houses tell us little about the immense costs that slavery imposed on enslaved people, or the wider US economy. Allen et al.'s results provide some numbers to quantify those costs, even if they cannot capture the full human toll.

[HT: Marginal Revolution]

Sunday, 8 November 2026

This week in research #151

Here's what caught my eye in research over the past week:

  • Dylong and Uebelmesser (open access) find that survey respondents in eastern Germany who receive information about the true size of the immigrant population in the region show more supportive attitudes toward immigration, with respondents who have less contact with immigrants prior to the intervention responding more strongly to the treatment
  • Luo and Sun find that days above 27°C are significantly associated with increased traffic accident incidence in Taipei, with larger effects for accidents caused by drivers of two‐wheel vehicles than those of four‐wheel vehicles
  • Duncan and Ross (open access) find, using a novel natural experiment based on the 2017 solar eclipse, that a 1 percent increase in Airbnb listings raises housing prices by 0.037-0.043 percent

Thursday, 5 November 2026

Book review: An Economic Theory of Greed, Love, Groups, and Networks

I just finished reading An Economic Theory of Greed, Love, Groups, and Networks by Paul Frijters with Gigi Foster, which has been sitting on my bookshelf for over ten years (and originally published in 2013), and kept being passed over in favour of other, seemingly more exciting, books. The book is clearly a labour of love for Frijters, and aims to add some of what he sees as the key missing parts of economics - not greed, which is inherent in the neoclassical model of individual utility maximisation, but love, groups, and networks. Frijters notes that the aim of the book is to add and integrate these into the mainstream view of economics, and:

What results is an enhanced view of how society works that increases the number of phenomena covered, and deepens the existing story.

In her preface to the book, Foster warns that the book may be an extremely uncomfortable read for those who hold mainstream economic views. She is right in that. I found it uncomfortable, because I was constantly thinking "yes, and...". The book has sections devoted to greed (the mainstream economic view, according to Frijters), love, groups and power, and networks and markets. The most uncomfortable aspect, I found, was that I thought the story they tell would have been much more compelling if told in the reverse order, starting with networks, and ending with greed and love.

There is a lot to like about the book. Frijters writes well, and tells a story that is sufficiently plausible to be persuasive. The problem is that much of the book reads like a collection of just-so stories, plausible enough individually and collectively, but often difficult to distinguish empirically from alternative explanations. I don't often find myself using the word epistemology, but this time I can't really avoid it. I don't buy into Frijters's epistemological stance.

As one example, Frijters conceptualises love in the form of loyalty to others. He dismisses altruism as a more limited concept than love. However, I was surprised by how little attention he pays to the broader literature on social preferences, particularly negative social preferences. People don't just care about others in a positive way. We can be spiteful, or envious. It is therefore rather ironic that later in the book, Frijters critiques the work of Akerlof and Kranton (whose book Identity Economics I reviewed here) as being "rather limited". I would say that a theory that considers positive social preferences, but gives little attention to negative social preferences, is rather limited as well.

Where I found the book was strongest was on the treatment of networks and markets (which is why, I think, it should have come first). I especially liked the way that Frijters leverages networks into explanations of the construction of cliques, groups, and ultimately, nations. The different forms of groups were also interesting, but I found this part of the book less convincing. I expect that there are many other categorisations of groups that could be developed, and I would have liked to see deeper engagement with the extensive literature on collective action, such as the work of Mancur Olson or Nobel Prize winner Elinor Ostrom. In her preface, Foster asks readers to forgive instances where particular authors' contributions are not acknowledged. However, I think the lack of engagement with the collective action literature is a more substantial omission.

The book is mostly heavy going, but there are some lighter parts. I particularly liked this bit:

The type of economist who depicts the rest of the world as being inhabited by agents who are rational and have complete information is, by implication, redundant. Instead of working out how the economic system works, such an economist could simply walk up to the first person on the street and ask him how the economy works...

In that, Frijters is undoubtedly correct. Overall, I think the book presents some interesting points to think about, things that could expand the horizons of economic theory. However, it's unclear to me who the intended audience for this book is (or was). I don't think it will engage the general reader, and a specialist economist may well struggle, as I did, with the epistemological approach. Sadly, it's not a book that I would recommend for any but those who are looking at extended mainstream theoretical economic models, and even then, they would likely be better served by looking at Frijters's more focused academic research.

Wednesday, 4 November 2026

New evidence that declining fertility rates may not stabilise

It is not news that total fertility rates across most countries are declining, and have been for some time (see the links at the end of this post). However, in spite of declining fertility, many population projections assume that fertility decline will come to an end, or even reverse. For example, the UN World Population Prospects projections assume that fertility rates eventually enter a relatively stable phase, with an average long-run level of about 1.6 births per woman, although the assumed stabilisation level varies across countries (see page 29 of this methodological report). Similarly, Stats NZ assumes that the median period total fertility rate will slowly decline from 1.59 in 2025 to 1.55 in 2054 and beyond (see here).

Given that it has been relatively rare for total fertility rates to reverse trajectory, how realistic is it that they will attain some stable equilibrium in the future? This new NBER Working Paper by Diane Coffey, Michael Geruso, Dean Spears (all University of Texas at Austin), and Sangita Vyas (City University of New York), digs into the data to provide an answer. Specifically, they look at data on:

...period total fertility rates (TFRs) for all country-years since 1950 in the United Nations World Population Prospects, in completed cohort fertility for country-cohorts in the Human Fertility Database, and in fertility measures from subnational geographies in India, Japan, and the United States.

It is worth taking a moment to explain the difference between a period total fertility rate and a cohort total fertility rate. Most of the time, the statistics we see reported are period total fertility rates. Those are the number of births a woman would have across their lifetime if they experienced the age-specific fertility rate at each age in that year. Of course, each woman only experiences the age-specific fertility rate of one age in any one year, not all of them collectively. That brings us to the cohort total fertility rate, which is the average number of births each age-cohort of women (born in the same year) actually have across their lifetime. Completed cohort fertility is computed after an age cohort of women has effectively moved out of childbearing age, which is why this statistic isn't reported often - you have to wait until women are over 50 years old before you can be confident that you have an accurate measure of the cohort total fertility rate. This distinction between the two measures matters because period fertility rates can decline if women delay having children, even though the cohort fertility rate (the average number of children women have over their childbearing years) remains unchanged.

Now, coming back to the working paper by Coffey et al., they use phase diagrams to illustrate the trajectory of total fertility rates over time. A phase diagram graphs the change in a variable against the level of the variable. It essentially shows, for each level of the variable, whether the variable is increasing, or decreasing. If a total fertility rate is to stabilise at some level, then the phase diagram should show that for low levels of the rate, the change is positive, while for high levels of the rate, the change is negative. The level where the change is equal to zero is the level at which the rate would stabilise. In other words, for fertility rates to stabilise at 1.6 births per woman, we would expect them to increase if the rate is below 1.6 and decrease if the rate is above 1.6.

What do Coffey et al. find? The results for past data from the UN World Population Prospects is summarised in Figure 3 from the paper:

Notice that for most levels of the total fertility rate, the point estimate of the change (growth) in the rate is negative. The most charitable interpretation of these results is that maybe there is some weak evidence for a cross-over at a total fertility rate of between the 1.25-1.35 bin and the 1.35-1.45 bin. However, none of the estimates below the 1.45-1.55 bin are statistically significantly different from zero, and the point estimates at the lowest total fertility rates are negative.

Moving onto data from the Human Fertility Database, the results are summarised in Figure 4 from the paper:

In this case, there is no evidence to support a stabilisation of cohort total fertility rates in these data. All point estimates are negative, and all of those from the 1.45-1.55 bin and above, are statistically significantly below zero. The results for Japanese prefectures since 2010, US states since 2010, and Indian districts (based on data from the Demographic and Health Survey 2019-2021) all similarly show negative and statistically significant change in total fertility rates across almost all levels of observed total fertility rates (the one exception is the 1.05-1.15 bin for Japanese prefectures, but that result is positive but statistically insignificant).

Taken altogether, these results demonstrate that there is no evidence that declining fertility rates will reverse course, at least not on average across countries, or across subnational areas in the three countries Coffey et al. looked at. There isn't even any evidence that total fertility rates will stabilise, at least not at the levels that have been observed so far.

Low and declining total fertility rates are the norm across developed countries, and increasingly in developing countries as well. They aren't returning to replacement levels. That doesn't mean that fertility rates will decline indefinitely, but it does mean that assuming they will stabilise seems unlikely to be borne out. So, policymakers might be better off planning for continuing low fertility, rather than blaming pet ownership and waiting for a rebound.

[HT: Marginal Revolution]

Read more:

Tuesday, 3 November 2026

Pets and children may be complements, not substitutes

When you see people treating their fur-babies like children, you might be tempted to wonder to what extent people substitute pet ownership for raising children. You wouldn't be alone in considering the relationship between pet ownership and raising children. In 2022, Pope Francis decried the low global birthrate, noting that "dogs and cats take the place of children". But are dogs and cats really taking the place of children?

A new working paper by Kuan-Ming Chen, Ming-Jen Lin (both National Taiwan University), Hau-Hung Yang (Academia Sinica), and Shirley Yu-Hsuan Yen (National Taipei University) provides an answer, using data from Taiwan. They take advantage of dog and cat registry data from 2009 to 2020, matched to tax administration records that identify families and household composition.

To work out whether two goods (in this case, children and pets) are substitutes or complements, we need to look at the cross-price elasticity of demand, which relates the percentage change in quantity demanded of one good to a change in the price of another good. When the cross-price elasticity is a positive number, the two goods are substitutes. When the cross-price elasticity is a negative number, the two goods are complements.

Now, there are two interesting things about this paper that would be familiar to intermediate microeconomics students. First, Chen et al. are careful to distinguish between Marshallian cross-price elasticities (named after late 19th Century economist Alfred Marshall) and Hicksian (compensated) cross-price elasticities (named after 1972 Nobel Prize winner John Hicks). The difference between the two is that Marshallian cross-price elasticities combine both the income and substitution effects, while Hicksian cross-price elasticities look at only the substitution effect (see this post for more on the difference between an income effect and a substitution effect).

The second interesting thing is that Chen et al. effectively use a Slutsky decomposition to identify the cross-price elasticity of children to a change in the price of pets. This essentially involves two equations. The first equation is Slutsky symmetry:

The left-hand-side variables are the share of household spending on pets (spet) multiplied by the Marshallian cross-price elasticity of pets to a change in the 'price' of children. The first term on the right-hand side is the share of household spending on children (schild) multiplied by the Marshallian cross-price elasticity of children to a change in the 'price' of pets. The second term on the right-hand side multiplies the two shares of household spending by the difference in the income elasticity of demand for children and pets.

The second equation is much simpler. The Hicksian cross-price elasticity of children to a change in the price of pets is equal to the equivalent Marshallian cross-price elasticity plus the share of spending on pets multiplied by the income elasticity of demand for children. In other words, this equation adjusts the Marshallian elasticity to remove the income effect, leaving just the substitution effect.

So, the Hicksian cross-price elasticity of children to a change in the price of pets can be calculated using information about all of the other variables in those two equations. And gathering estimates of those other variables is what Chen et al. set about doing.

First, to estimate the cross-price elasticity of pets to a change in the 'price' of children, Chen et al. look at how increases in the 'child bonus' paid by local municipalities affect the registration of pets, since the child bonus affects the up-front cost (or 'price') of having children. Using an event study, they find that:

Both cats and dogs respond, and they respond in an ordered way. In the first post-treatment year, new dog acquisitions rise by 0.067 percentage points on a pre-treatment base of 0.332 percentage points, a 20.1% increase (𝑡 = 2.40); new cat acquisitions rise by 0.0136 percentage points on a base of 0.125 percentage points, a 10.9% increase (𝑡 = 2.36). Cats are therefore complements with children as well, but the proportional response is roughly half as large as for dogs.

...the implied Marshallian cross-price elasticity for cats is approximately −0.68, against −1.24 for dogs.

Remember that the Marshallian cross-price elasticity doesn't remove the income effect of the price change. To calculate the Hicksian cross-price elasticities, Chen et al. therefore need to estimate the other parts of the equation. So, next they estimate the income elasticities of demand for children and pets, making use of an interesting lottery:

Taiwan’s sales tax system provides a natural experiment for income shocks through its uniform invoice lottery program. As part of a nationwide effort to reduce tax evasion, nearly all consumption receipts are eligible for bi-monthly lottery drawings, with prizes reaching up to NTD10 million. Receipt holders are randomly selected, and prizes are distributed regardless of income level or background. These awards thus constitute plausibly exogenous, positive income shocks.

This lottery had already been used by other researchers in an event study (where the treatment is receiving a lottery windfall of at least NTD2 million) to estimate the income elasticity of children, finding it to be 0.15, meaning that children are a normal good. Following the same approach, but looking at the income elasticity of pets, Chen et al. find that the income elasticity of dogs is 0.11. Annoyingly, they don't report the income elasticity of cats (although they use it in their calculations later), although their other estimates suggest that it may be close to zero.

Finally, Chen et al. estimate the share of expenditure on children and pets. Using data from the Family Income and Expenditure Survey, they find the share of household spending on pets is 0.032 (about 3.2 percent). For children, they use an 'equivalence scale approach', which weights children in adult-equivalent terms at 0.3, and noting that would imply a share of spending in a three-person household of about 17 percent, they adopt a conservative share of household spending on children of 0.2.

Now, with all of the pieces together, Chen et al. move on to estimating the Hicksian cross-price elasticity, finding that it is equal to -0.198 for dogs, and -0.089 for cats. In other words, dogs and children are complements, and so are cats and children. The results are also reasonably robust to various alternative assumptions about the income elasticities and expenditure shares.

Chen et al. then go on to explore why pets and children might be complements. By looking at the timing of pet ownership and childbearing, they identify that it is much more common for families to have pets first, and children second. They infer from this that:

Pets may serve as a “starter family,” letting prospective parents rehearse caregiving and learn about their own taste for family life before a birth.

Of course, the mechanism results do not eliminate other explanations. The results are consistent with the 'starter family' channel, but it is also possible that people who are already planning to have children are more likely to register a pet beforehand. Also, these results are based on Taiwan. Perhaps pets and children are substitutes in other countries - we won't know unless further research is done. Nevertheless, these results are a good place to start, and should make us really question whether dogs and cats are taking the place of children.

[HT: Jacques Poot]

Monday, 2 November 2026

William Nordhaus, 1941-2026

I was saddened to hear of the passing of William Nordhaus last Friday. Nordhaus won the Nobel Prize in 2018 "for integrating climate change into long-run macroeconomic analysis", and it is his work on climate change and economics that he might be best known for. However, that was far from his only contribution to economics. My ECONS102 students might recognise Nordhaus as one of the first economists to formally analyse the tradeoff between strong intellectual property rights, which provide incentives for creation, but lead to under-consumption, and weak intellectual property rights, which can lead to optimal consumption, but lead to under-investment in idea creation (for more on that tradeoff, see this post).

My favourite Nordhaus contribution was actually some work he did on the change in the cost of light over time. The original research paper is available here, but the story is best told in this article from the BBC:

Back in the mid-1990s, an economist called William Nordhaus conducted a series of simple experiments with light.

First, he used a prehistoric technology: he lit a wood fire.

But Prof Nordhaus also had a piece of hi-tech equipment with him - a Minolta light meter.

He burned 20lb (9kg) of wood, kept track of how long it burned for and carefully recorded the dim, flickering firelight with his meter.

Next, he bought a Roman oil lamp, fitted it with a wick, and filled it with cold-pressed sesame oil.

He lit the lamp and watched the oil burn down, again using the light meter to measure its soft, even glow.

Bill Nordhaus's open wood fire had burned for just three hours on 9kg of wood.

But a mere eggcup of oil burned all day, and more brightly and controllably.

Why did he do this?

He wanted to understand the economic significance of the light bulb.

His research showed just how dramatic the fall in the cost of light was. Drawing on his work, the same BBC article notes that 60 hours of labour:

...had once produced the equivalent of 54 minutes of quality light [but] now produced 52 years.

That fact demonstrates just how powerful technological progress and economic growth have been over the very long run. But Nordhaus's deeper insight was that simply looking at prices and economic growth alone can lead us to miss the enormous improvements in living standards. The significance of that cannot be overstated.

I'm surprised that there doesn't appear to have been much coverage of Nordhaus's passing. Nevertheless, Yale News has a good obituary. And while searching for some others, I ran across this podcast from earlier last month, that neatly covers his life in the way a good obituary would.

Nordhaus's work extended far beyond the climate economics he is most known for. His research on the cost of light is a reminder that seemingly ordinary things can nevertheless tell important stories. He will be missed.

[Update: New York Time obituary was published on 2 November (ungated here), shortly after my post, and see also this excellent piece by Nordhaus's nephew Ted]