More people, more ideas – in the long run

More people means more ideas. This concept underlies arguments ranging from Julian Simon’s belief that human living conditions will continue to improve through to Bryan Caplan’s argument that we should have more kids. While I don’t always take this concept to the extent of Simon or Caplan (as I have posted on before), the concept must be right at some level. One person will have more ideas than zero people. One hundred people will have more than one person. You can argue about diminishing returns and so on, but the basic concept must hold.

For me, some of the more interesting evidence is over the long-run, which Michael Kremer discusses in a paper titled Population Growth and Technological Change: One Million B.C. to 1990. Kremer bases his argument on the Malthusian concept that population is a measure of technology. In a Malthusian state, the environment and level of technology constrain population. As technology grows, a given area can support a higher population, so technological progress is directly linked to population growth.

Kremer showed that population growth has accelerated over the last million years (although this has slowed the last hundred or so), with population growth faster than exponential growth. This is consistent with the idea that as the population grew, people generated ideas faster and faster, further accelerating the technological growth rate, and hence population growth. The following diagram shows Kremer’s point, with the population growth rate increasing with population size until recent times.

Kremer  created a number of models to show this point. In one model, he included the concept that research productivity increases with income, which might explain why some large populous countries (China and India) have lower levels of technological development (for the moment). Kremer showed that even though this creates some ambiguity about the relationship between population and technological growth, the concept that larger populations have higher levels of technological progress held.

Another concept addressed by Kremer is that productivity might depend on the size of the population. This encompasses positive network effects whereby more people allows more specialisation and service of a larger market, and negative effects such as “stepping on toes” whereby researchers duplicate each others’ efforts. Including these effects in his model did not change the basic finding that technological progress increases roughly in proportion with population.

Kremer tested the model predictions with some basic regressions against long-term population data. Not surprisingly, they all showed a strong correlation between population size and population growth.

The paper gets more interesting with some of Kremer’s analysis of inter-regional differences. Taking five successively smaller populations: the old world, the Americas, Australia, Tasmania and Flinders Island. For each of these areas, population density increases with land area, suggesting that higher population are able to sustain a higher level of technology. Kremer writes:

As the model predicts, in 1500, just after Columbus’ voyage reestablished technological contact, the region with the greatest land area, the Old World, had the highest technological level. The Americas followed, with the agriculture, cities, and elaborate calendars of the Aztec and Mayan civilizations. Mainland Australia was third, with a population of hunters and gatherers. Tasmania, an island slightly smaller than Ireland, lacked even such mainland Australian technologies as the boomerang, fire-making, the spear-thrower, polished stone tools, stone tools with handles, and bone tools, such as needles [Diamond, 1993]. Flinders Island, near Tasmania, has only about 680 square kilometers of land, and according to radiocarbon evidence, its last inhabitants died out about 4000 years after they were cut off by the rising seas-suggesting possible technological regress.

In some ways, the general message of Kremer’s paper appears obvious. However, it is one of those papers where the feedback relationship between population and technology makes it difficult to confirm the direction of causation (even though I agree with the general concept). Regardless of the source of technological progress, the Malthusian model predicts that population will increase to match it. Higher population levels will always have a higher level of technology, regardless of the source of innovation.

The element in the data which lends strength to Kremer’s argument is that technological progress accelerates when there is a larger population. However, the length of time over which Kremer makes his observations raises an important question. Human ancestors only achieved our modern brain size in the last 100,000 or so years (peaking 30,000 years ago). There is evidence of a cultural leap forward around 50,000 years ago. The source of ideas is a very different creature at each end of Kremer’s time series. How much of the acceleration is generated by greater population, and how much by a more productive population?

4 thoughts on “More people, more ideas – in the long run

  1. This reminds me of the thesis of Geoffrey West and colleagues that things like patents and other measures of innovation increase superlinearly with city population growth, so that doubling the size of a city more than doubles innovation. If I remember correctly West sees this being driven by scaling up of inter-personal networks, so maybe a similar phenomenon could occur on a regional or continental scale (e.g., as in the comparison of Asia to the Americas to Australia to Tasmania) even before cities existed.

  2. Biologists and social scientists generally believe that population growth represents expansion into unexploited resources, i.e., increasing towards the carrying capacity. However, it is probably better to assume that at every stage the population is already at the carrying capacity and that increasing numbers represent a gradual increase in the carrying capacity itself, due to accumulated inventions.

  3. Increases in carrying capacity due to inventions must be seen as temporary.   Ideas alone do not run a high tech infrastructure, it requires trained, intelligent, innovative, competent people with at least a modicum of integrity and impulse control / executive function.   Such populations represent a store of human capital, which increases with greater numbers as long as those characteristics persist.

    If  such populations of high human capital are replaced by populations with lower human capital, then the high tech infrastructure cannot be sustained below a certain threshold — regardless of the numbers of residents.

    Populations are not fungible, despite the somewhat lucky experience of the US with immigration up to this point.   Different populations have different evolutionary histories which have endowed them with a greater or lesser quantity of genetic and cultural human capital.

  4. Kremer is right as far as he goes–and he makes a point that is so often ignored: that more people means more ideas–so we all get the benefits from there being more geniuses.

    So good on him for that.

    But this only one part of a much wider and much more important point: that more people also means a much increased division of labour, for which the benefits from genius are but one aspect–not least of which is increased specialisation and concentration on areas of greater comparative advantage; the greater multiplication of knowledge, and ability to produce products that would otherwise be impossible; greater economies of learning and motion that increase with increased specialisation; greater gains from new machinery attributable to specialisation under division of labour.

    And not just more people, but increased population density–which means lower communication and transport costs and increased specialisation. Even Plato recognised the increased specialisation consequent on the growth of cities.

    And the point about Plato suggests a further point: that while it is great that arguments like Kremer’s are being given currency–and more power to him for that–it is astonishing how little recognition is given to earlier thinkers who not only developed this idea, but more importantly the wider implications of the benefits of both division of labour and increased population density.

    Preeminent on this score especially are thinkers like Adam Smith and the too-often overlooked Frederic Bastiat–who are virtually unique in teaching (in contrast to Malthus) that population growth is a net benefit to everyone. Smith of course began his famous book by explaining the greater productivity arising from increasing division of labour, observing that in a division-of labour society a larger population means a greater and more intensive division of labour. This is what he means in Book 1, Chapter 3 of ‘Wealth of Nations’ when he writes “the division of labour is limited by the extent of the market.”

    Bastiat, in his posthumously-published book ‘Economic Harmonies,’ formulates the admirable principle in his chapter 4, on ‘Exchange’:
    “Other things being equal, an increase in the density of the population means an increase in productive capacity.”
    The principle is further teased out by his translator, W. Hayden Boyers, from notes left by Bastiat, and it is here in a five-page exegesis that the fullest Classical expression of the principle can be found. [You can read it online as Note 40 in the ‘Population’ chapter of this translation: ]

    Unfortunately, however, both Smith and Bastiat’s positive points are marred by their reliance on the failed “labour theory of value.” I would therefore draw the attention of interested readers to the integration of the work of both thinkers on this point by a contemporary economist, George Reisman, (whose treatise on economics can be downloaded at

    Reisman’s section on “Economic Competition” from pages 343 to 371 is I think the best summary of the argument of “the general gain from the existence of others,” of which the benefits of increasing population density is a particular case; Reisman lays out the framework for this principle very clearly, making plain the points of both Smith and Bastiat on the basis of modern value theory–making it abundantly evident (without mentioning Kremer explicitly) that Kremer’s point should nonetheless be seen as part of a much wider argument, with roots in the thinking of these two seminal economic thinkers.

Comments welcome

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s