Things to think about #7 - Glenn Loury, Life History Theory, LatAm fertility and Substack economics
I am a big fan of US economist and academic Glenn Loury. He is smart, honest and well-articulated. He is also not afraid of an intellectual scrap if he stumbles upon one. He is an indispensable public commentator and intellectual whose ideas and influence go far beyond the confines of race, and associated social issues, in the US where he has staked his claim to fame and authority most comprehensively. Glenn has an impressive back-catalogue of writing and citations, but the best way to get a sense of him is by listening to his podcast the Glenn Show, which can be found on all the usual platforms. I am also a big fan of his co-conspirator, John McWorter, a US linguist and public intellectual, with whom Glenn runs a bi-weekly conversation on his podcast, and Q&A for paying subscribers. It is a must-listen. On this occasion, however, I want to recommend Glenn’s recent discussion with Larry Kotlikoff, a US academic economist, in which they discuss the economic policy ideas of the two candidates in the upcoming US presidential election ideas, and the US economy more generally. As the title of the podcast goes; if only we had an economist in the White House!
Life History theory
I am currently tidying up some of my writing on demographics after finishing the fifth chapter in my long-running project to explore the drivers and consequences of global demographic shifts. As part of this process, which includes re-writing some of my work to make it suitable for “audification” (see the first example here), I re-read some papers on Life History Theory, LHT, and in particular, works by Ruth Mace, professor of evolutionary anthropology at UCL. Mace’s work remains one of the best gateways into understanding the fundamental drivers of fertility in so-called natural fertility societies, and how such drivers have shifted, and may have become warped, by modernity. This is the question of how to explain, with an evolutionary lens, why the correlation between birth rates and rising wealth/income appears to be negative in a modern context, or at least not as straightforwardly positive as a simple evolutionary framework would suggest. From the perspective of LHT, this shift occurs because the inherent trade-offs for humans have shifted over time, and decisively in the context of the industrial revolution and modernity.
Kaplan and Gangstead (2004) describes LHT as follows:
LHT provides a framework that addresses how, in the face of tradeoffs, organisms allocate time and energy to tasks and traits in a way that maximizes their fitness. Optimal allocations vary across the life course and, hence, LHT generally concerns the evolutionary forces that shape the timing of life events involved in development, growth, reproduction, and aging.
I explain LHT in more detail in the fourth chapter of my demographics project. For an introduction on Ruth Mace’s work and LHT, this paper from 2000 and this paper from 2014 are good places to start. The following passage from the first is a good summary of the key idea.
Life history theory is a well-developed sub-discipline of evolutionary ecology which is explicitly concerned with the timing of life history events (growth, reproduction, maintenance, and death) under natural selection. The timing and the scheduling of births in order to maximise fitness (in the Darwinian sense) can in theory be calculated for any given environment and any given set of environmental constraints. In examining the timing of births, researchers often try to understand both the behavioural and the physiological determinants of fertility (e.g., menopause). Parental investment is viewed as a life history trait that is determined in part by considerations regarding the quantity versus the “quality” of offspring. It has long been recognised that maximising reproductive success is not necessarily about maximising fertility alone. A “Darwinian demon” who reproduces at the maximum rate would not succeed in the real world, as there would be costs of reproduction for the mother (and probably even for the father), and there would be competition between siblings for parental resources. Trading off these costs with the fitness benefits of fertility ultimately determines the rate and nature of human reproductive decisions.
Fertility in South America
X-account Birth Gauge posts a very nice table of the shifts in global births and total fertility rates across countries, from 2015 to 2024. Not surprisingly, the trend is down virtually everywhere. Benjamin Wolf picks up on the rapid decline in four South American countries; Chile Colombia, Costa Rica and Mexico. So, what’s going on specifically in these economies? Two things; first a strong quantum effect seen via falling births across all age groups, and secondly, the emergence of a bi-modal distribution as a function of women’s educational attainment. Lima et al. (2018) have the key data and charts, and this is how I explain it in my chapter on the tempo effect of fertility:
Fertility in many Latin American countries fell below the replacement level in the early 2000s, indicating a shift along the lines observed in other key regions. But there is a twist to the story in Latin America. The decline is due almost exclusively to quantum effects. The number of births to women in their peak reproductive age stood at 270 per 1000 women in 1970, only down slightly from 1950. By 1990, it had declined to just under 180, and it fell further in the five years ending 2020, to just under 100. This is consistent with evidence in other regions, but LatAm stands out in terms of tempo effects. Indeed, the age of the peak crude birth fell since 1950, from 25 to 22-to-24 in the early 2000s. It was 24 in 2020, according to the UN. In other words, the decline in the region’s total fertility rate from around 6 in the early 1960s to just over 2 in 2010 was due exclusively to quantum effects. Lima et al. (2018) investigate the relative lack of birth postponement in Latin America, invoking the idea of a bi-modal fertility regime, split along the axis of women’s educational attainment. Young, and relatively un-educated, women are having their first child relatively early, and maintain relatively high levels of fertility through their early reproductive years. By contrast, women with tertiary education are now engaged in significant postponement, in line with evidence in developed economies. Lima et al. (2018) provide convincing evidence of this hypothesis via early 2000s consensus data from Chile and Brazil. In Chile, the age of first birth for women with primary and secondary education cluster in the early 20s, while it jumps to the early 30s for women tertiary education. This gap is even larger in Brazil, with the age of first birth for women with tertiary education is in the mid-30s.
The following chart from Lima et al. (2018) shows this phenomenon for Chile and Brazil.
Figure 7. P. 16 in “Twin Peaks: The Emergence of Bimodal Fertility Profiles in Latin America”, Vienna Institute of Demography Working Papers, No. 10/2017
Such bi-modality by which women with tertiary education opt for birth postponement to a significantly larger degree than women with only primary or secondary education is likely a key driver of what has been a very rapid decline in fertility in so-called developing and emerging economies since the beginning of 2000s. This is because the current generation of reproductive and working-age women in these economies are now, in many cases, the first who are able to obtain tertiary education compared to their mothers and grand-mothers. A rapid tempo-effect driven decline in period fertility is exactly what we would expect in such a situation.
Substack economics
I am gravitating closer to Substack these days, as a consumer, but I am still hesitant to jump in with both feet. I should be very excited about the platform, and in a way, I am. I don’t have the claim to be an early adopter of much, but when it comes to (econ)blogging, I can make a pretty good case that I was one of the founding members of the fledgling econ-blogging community who emerged in the primordial ooze of Web 2.0. Alpha Sources started in 2005, which means that it will soon celebrate its 20th anniversary. Not many blogs can say the same, I think.
In the early days, the idea of selling blog content as a newsletter was unthinkable, mainly because that was what, well, newsletters did. This is to say, if anyone did this at all. Some blogs experimented with ad-revenue early, but it was pretty obvious that only a few blogs could hope to generate anything resembling meaningful revenue. I still think that’s the case, though there will undoubtedly now be a good number of newsletter and blog sites that are able to pull in good money from ad revenue in the significantly larger and deeper internet compared to the early days of Web 2.0. Whether they’re worth reading on a screen is another question entirely.
As a writer, it warms my heart to see a platform like Substack empowering other writers and content creators to carve out a niche for themselves in anything from brand watching, single stock picking, economics to cultural comment and critique, just to name a few topics. Many Substack writers use the platform as a supplement for their wider work and reach, rather than as the sole platform for their distribution. But that doesn’t invalidate the underlying point. In its simplest form, Substack empowers the individual writer to capture a following, and if they’re sufficiently patient, to make a living. This is the theory.
In reality, however, you only have to scratch a little bit in the surface to realise that on Substack, just as in other walks of life, the Matthews Principle applies. A few writers on Substack are doing very well, while the rest are earning pennies. I’d be interested here in data on whether there is a relatively successful mid tier who are able to extract meaningful revenue. I doubt it, but I would be happily proven wrong. Success will always be top-heavy on such a platform despite Substack’s best effort to use its scale to allow users to receive targeted recommendations for content that there is a good chance they’d be willing to pay for.
If the economics of Substack publications at large are questionable, the economics for the user are more uniquely poor, and this is a big problem for the platform. I love the idea of using Substack as my media platform of choice in a world where traditional media offer one disappointment after the other. What I don’t love is that it will cost me anywhere from $100-to-150 per month to build and curate a truly powerful and useful Substack experience, and this is if I hold myself back. Worth it? Maybe, but considering the bundle of traditional and even non-traditional media subscriptions I can get for that amount of money, the competition is pretty stiff.
This then begs the obvious question. Could a Substack-style service work on a Spotify/all-you-can model, where users pay a fixed monthly fee, say $20 per month, and get access to all content? I think it could, and I have a spreadsheet model to prove it—you’ll just have to trust me on this—but not without modifications. The main and significant modification is that you have to curate the authors on the service, and only bring on new authors using a carefully calibrated algorithm in line with rising revenue. This sounds complicated, but the quantitative aspect of this really isn’t. The key is how to choose to best/correct authors, and which niches and content to on-board first, second, third etc. This is an art, not a science. I know what many people might say here; congratulations you’ve invented a newspaper, or a magazine. But I really haven’t. The model I propose is one with no editorial oversight, where each writer has his/her own platform embedded in a larger brand. It is a platform where the user, like now on Substack, chooses his/her own content setup, independent of what others are doing.
How much would writers get paid? There are many models, but I believe such a service would be able to pay out a significant chunk of revenue to authors—60-to-70%—while still retaining strong bottom-line margins. Substack currently takes a 10% cut of all single subscription fees, though it is possible that bespoke deals now exist between Substack and its most prolific writers. My model would take a bigger share of revenue, but also pay out much more to its authors, by definition as there would be control on how many authors would be on the platform as a function of revenue. The compensation paid to authors would be split between an equal share to everyone and a bonus share to the most popular authors. You would also need to maintain a relegation trapdoor for authors with no readership, or more likely, for authors who are not abiding to the minimum demand for published quantity, and perhaps quality too. More generally, the platform would need to retain the exclusive ability to dump an author if it so pleases—not sure whether this is the case on Substack currently—but not necessarily in my view, exclusivity on the content itself. This former right would have to be used with extreme caution. My sense is that Substack is a staunch defender of its authors to have a very range of views on the platform. I am a big supporter of this position.
Could Substack in its current form introduce elements of an all-you-can-eat model? I guess, most obviously via some form of bundling. Will it? I am not holding my breath. The big question for an all-you-can-eat written-content service, with independent writers, is whether such a model would scale beyond what similar multiple-author blogs, newsletters etc already are doing? I think it could over time, but the proof of the pudding is in the tasting.