A Distillation of the Rise and Decline of Nations
Mancur Olson (MO)’s book “Rise and Decline of Nations” has been on my reading list for a while; I finally got around to reading it. I’ll give a one-paragraph summary, before delving into more details.
The book’s key point is that “distributional coalitions” will naturally emerge over time in stable societies (no revolutions) with unchanged boundaries (no significant expansion of territory). Those coalitions will find it in their interest to lobby for policies that will benefit them to the detriment of societal efficiency. Those policies will unfairly redistribute wealth to those coalitions: while they shrink society’s overall production frontier, they increase the benefiting coalitions’ share to a greater absolute level than when the production frontier was optimal. Therefore, the longer a society has been without revolutions or territorial expansions, the more it will diverge from its production frontier, resulting in decline.
That’s the tl;dr. If you want to dive a little further, read on.
The difficulty of collective action
MO’s key argument around the emergence of such “distributional coalitions” starts off with a discussion of the concept of collective goods. Collective goods are those goods that, once produced, go to everyone in some group or category. The producer of the good cannot selectively exclude individuals in that group or category from enjoying the benefits of that good. Public goods are a special case of collective goods that benefit the whole of society (e.g., law and order, defense, lack of pollution). Individually, we have an incentive for someone else to contribute to the production of those goods, since the producer cannot exclude a non-contributor from the benefits. This is the reason free markets aren’t able to supply public goods, and why we rely on state coercion via compulsory taxation. Taxation is an example of what MO calls “selective incentives”, aka incentives –positive or negative– to induce beneficiaries of collective goods to contribute to the cost of their provision. Compulsory taxation is a case of negative selective incentive (state violence resulting in your imprisonment).
Importantly, there are other types of collective goods that are not public, meaning they only benefit a subset of the overall population, yet still characterized by non-excludability within that group. Examples of such organizations are trade unions, labor unions, lobbies, cartels, etc. A labor union negotiates for all workers, whether they pay dues or not. Such organizations have also developed both negative selective incentives (in the form of overt or covert coercion, or even shame) and positive selective incentives (provide some side excludable benefit to contributors).
Are there circumstances in which a collective good can be provided (“collective action”) without having to rely on selective incentives? The answer is yes. Collective action will happen, MO argues, when the group that would benefit from collective action is sufficiently small and the cost-benefit ratio of collective action for the group sufficiently favorable. If a collective good of value X costs Y < X to provide to a group of size N, then the smaller N, the more likely collective action is to take place, whether through successful bargaining amongst the group (about how to “fund” Y) or unilateral action (when X/N > Y). This still holds, albeit to a lesser degree, for larger groups that are federations of smaller groups.
All in all, small groups and/or groups with access to selective incentives will be more likely to have collective action. However, even then, collective action is no guarantee, and still has to face up to the fundamental problem of the individual incentive to free-ride. Successful bargaining or successful development of selective incentive structures will take time, and will only emerge in a stable environment, where enough repeated games (in the game-theoretic sense) can be played to give cooperation a chance. This is the main reason why collective action needs long periods of stability to start appearing in a society.
Of pubs and clubs
The next step in MO’s overall argument is that not all groups who could benefit from collective action are equal: some will be more likely to achieve it (be “better organized”) precisely because they have easier access to selective incentives or because they are smaller. Groups like “consumers”, “taxpayers”, “the poor”, “the unemployed” will typically be less organized than firm cartels or trade unions. And as MO puts it, “it would be in the interest of those groups that are organized to increase their own gains by whatever means possible. This would include choosing policies that, though inefficient for the society as a whole, were advantageous for the organized groups because the costs of the policies fell disproportionately on the unorganized.”
The fact that we can’t achieve symmetrical organization of all possible groups that have a common interest implies that any redistribution the better organized groups achieve through the political system will be suboptimal, meaning it will come at the expense of society in general. This asymmetry is why we can’t reach an optimal outcome when all those groups bargain over “the rules of the game” (legislation through the political process influenced by organized groups with common interest); and those rules of the game are what ultimately determine a society’s efficiency frontier. This is why MO calls those well-organized groups distributional coalitions, inasmuch as, at the end of the day, they redistribute wealth to themselves at the expense of society. This is truer the smaller the distributional coalition, since it can more easily externalize the negative efficiency effects and more easily concentrate the benefits of self-serving. The more encompassing a distributional coalition, the more its interest is tied to that of society’s at large, and so the more difficult to externalize the efficiency costs. Unfortunately, small groups are much easier to organize, so that you end up with a bunch of “clubs and pubs” slowly rendering the body politic sclerotic.
At this point, you might think that we need to strive to have encompassing distributional coalitions. MO, though, delivers the bad news: the fact that information about public goods is itself a public good –and therefore undersupplied– means that, even where encompassing organizations emerge, they might not know the right policies to pursue. As a side note, I do wonder, though, if this is true in practice. This would mean that everyone in these organizations is to some extent clueless. But is that realistic? Don’t people have a desire to be “lovely”, as Adam Smith would put it? Isn’t credible education signaling a counterforce to rational ignorance in politics? There’s some evidence that prospective mates select for public-spiritedness, thereby creating a positive selective incentive to be informed, partially privatizing knowledge about public goods via the mating market.
Notwithstanding, pursuing MO’s line of thought, in the end, we’re stuck between a rock and a hard place: on the one hand, we have encompassing organizations cluelessly pursuing public goods with no competition with other organizations and therefore no discovery process; on the other hand, we have many competing distributional coalitions with their narrow attempt at redistributing for a narrow constituency at a greater social cost.
A common criticism of this “distributional coalitions are all bad” take is that laissez-faire (within a minimal institutional context) –as an alternative to a world saturated with such coalitions– would lead to more inequality. MO disagrees, arguing that “there is greater inequality in the opportunity to create distributional coalitions than there is in the inherent productive abilities of people.” Abilities are normally distributed, MO argues, and while there may be larger differences in the holdings of capital, those differences are nevertheless bounded by decreasing returns to capital. And as capital becomes abundant, the wages of the associated labor should go up. Distributional coalitions are the real culprits.
Side note: there might be an interesting link between MO’s theory and Hirschman’s Voice v Exit framework. If you lack a culture of well-informed encompassing distributional coalitions, then Exit, exemplified by Silicon Valley’s culture of self-disruption, might the only way to combat sclerosis.
Business cycles explained
MO’s crowning achievement in this book — and perhaps the reason it’s so famous and considered a classic — is to take this theory he just developed, and use it to explain business cycles in general (“rise and decline”), and depression and stagflation more specifically.
MO explains that business cycles are ultimately caused by unexpected technology-driven, labor-saving, deflationary shocks, which, at equilibrium, will propagate to adjacent sectors. If those do not adjust their prices properly because of distributional coalitions, then real output (level of production in terms of quantities) will be affected. And if production factors prices (eg, labor wages) do not adjust either, involuntary unemployment will ensue.
Why do distributional coalitions tend to not adjust their prices in response to deflationary shocks? Because it creates scarcity, which funnels value to them. They could create scarcity by controlling quantities instead of prices, but scarcity via quantities would mean deciding who is going to produce less (whether in terms of actual goods produced or in terms of hours worked), and negotiating that can be costly; whereas just saying “sell whatever you can at price X” is easier, and so is saying “the more senior, the higher paid” instead of bargaining over who is going to work less. Involuntary unemployment is therefore caused by those sticky prices, themselves a symptom of a cartelized economy. This is where MO’s theory starts completing Keynes’ incomplete “general” theory of employment, interest, and money. MO gives a microeconomic foundation to Keynes’ unsubstantiated assumption of “sticky prices”.
The existence of sectors with flexible prices (highly competitive sectors with few distributional coalitions) is no consolation as it is unrealistic to expect swathes of labor from the “fixed prices” sectors to migrate to this “flexprice” sector. And as the share of the “fixprice” sector in an economy grows (which will happen in tandem with the gradual takeover of distributional coalitions in a stable society), so will the likelihood that any deflationary shock will turn into macroeconomic malaise. In short, the longer the stability of a country, the more cartelized its economy, the more deflationary shocks will be absorbed via real output adjustments and involuntary unemployment instead of price adjustments. Crucially, this is also true when the economy is booming (aggregate demand is high): distributional coalitions, expecting inflation, will be incentivized to set their prices at non-clearing levels, leading to the infamous stagflation (inflation is high, yet unemployment is high too, because of the non-clearing prices). This is where MO reconciles monetarists’ rational expectations theory with Keynes’ sticky prices.
Armed with this new framework, MO gives an interesting take on Keynesianism as the belief that the government can outsmart the distributional coalitions by shoring up aggregate demand, thereby raising all prices, thereby bringing down the relative prices of the fixprice sector (assuming it cannot adapt its prices fast enough). Not a necessarily totally bad idea.
That’s basically it. The book has a ton more discussions of statistical evidence for the theory, and a ton more illustrative examples. The above, while somewhat dense and theoretic, should nonetheless be a fairly high-fidelity distillation of this classic book. If you’ve made it this far, ideal, intrepid, undeserved reader, I salute you, and welcome any feedback from you.
