Democracy, Decentralization, and Well-Being

Todd Simpson
10 min readDec 11, 2017

--

Those of us raised in democratic societies take it for granted that they provide better well being (for common individuals) than other forms of governance. At the heart of democracy is personal freedom and autonomy, backed by the rule of law.

We also take for granted the interplay of decentralized versus centralized authority. Decentralization can mean many things, but here we refer to it in terms of power, authority, and decision making. The more authority individuals have, the more decentralized the power system in which they are operating.

Almost by definition the more democratic a system, the more decentralized it is. Of course, authority can be too decentralized — “every person for themselves” — so we put limits on decentralization through that same rule of law.

With the advent of decentralizing technologies, which make possible more decentralized systems, it is worth doing a high level check on the assumption that decentralization / democracy actually leads to better well being. We will do so by simply plotting public data sets, and (spoiler) there are no surprises. We want to see if our societies are at the limit of decentralized structures, or whether there is room for further improvement. We want to see the trend lines.

To get us started, consider the following axis: In a fully centralized system, a single individual or small group is making decisions for everyone; In a fully decentralized system, each individual makes all the decisions for themselves. We can map almost any system — relative to other comparable systems — on this axis.

We start by using the Economist Intelligence Unit’s Democracy Index (ECU-DI) as a measure of the level of personal authority and autonomy within a nation. The index measures five variables: electoral process and pluralism, civil liberties, functioning of government, political participation, and political culture. It measures not only an individual’s ability to contribute to decision making, but the expectations and environment in which they operate. Thus, it is a very good indication of personal autonomy. The ECU-DI maps the resulting scores into four major categories, mapped onto our axis below.

Democracies, and decentralized user authority, fail without some strong centralized structures, especially the rule of law. So that explains why on the full range of this axis, all nation states tilt towards the right. In our case we are mostly concerned with the authority maintained by individuals, so centralized structures that promote individual freedoms can be seen as aids to decentralization, not as weighing the entire system towards a centralized structure.

It is possible to imagine a future where even the rule of law, and its enforcement, have more decentralized structures than they have today. Where instead of a hierarchy of courts, there are legal structures on a block chain with a mechanism for updating those structures by accredited participants. For the purposes of this paper, however, we will live with the dichotomy of strong central institutions enabling individual sovereignty.

To that end, we will use the ECU-DI as the base axis for judging ‘well being’. In the above diagram we have left lots of room to the left — there is a lot of space for more decentralized structures — but in the below we will assume that Norway is all the way to left on the graphs, and North Korea all the way to the right. The highest score in the ECU-DI is 10 and the lowest score is 0. In what follows we will use this range, and call it the D-Index — “D” for both democracy and decentralization.

Wealth and Decentralization

If we plot “Wealth per Adult” against the D-Index, we see that stronger democracies are correlated to higher personal wealth. While it is possible that more wealth makes it more likely for democracies to arise, it seems more likely that democratic structures have enabled more wealth generation.

What we also see in this trend line is that there is no indication of leveling off or reversal of the trend as you move to the far left — that is, we don’t see ‘too much democratization leading to lower wealth’. The obvious conclusion is that there is still room for improvement, beyond ‘Norway’. Even more decentralization may lead to even better wealth. While not conclusive, the graph also implies slightly better wealth for some highly centralized countries (those in the range 2.5 to 1.5) over those in the middle (from 6 to 2.5). The two main outliers are Iran and Saudi Arabia which can be explained by their natural resource wealth, and, as we will see, their poor wealth distribution — nevertheless, average wealth per adult is pulled higher.

If we look at wealth distribution, using the Gini index for wealth, we see that more decentralized societies have better wealth equality. A higher Gini score means higher inequality. Intuitively it makes sense that in authoritarian, highly centralized, countries, more relative wealth accumulates to the few as there are less checks and balances in the system.

There has been lots of discussion, highlighted by Piketty’s work, on the growth of inequality in modern democracies, which is certainly true. These data points represent a moment in time, and are fairly widely dispersed, with r² of only 0.107. Although there seems to be room for improvement through more decentralization, this may also point to the need for improvements in other structures to manage inequality. Piketty does a better job than we could hope to in analyzing those.

If we look at inequality over the last century in three democracies and one flawed democracy (the USA) Piketty’s data looks like this:

While not dramatic or conclusive, it is interesting to see that the three functioning democracies have performed better over the last 30 years than the USA. Speculatively this could be due to the USA being a duopolistic political environment, and with the weakening of antitrust laws in the Reagan era.

Unemployment and Decentralization

We can also plot unemployment against the ECU-DI, and also see the expected trend line. Higher democracy scores correlate reasonably well (r² of -0.165) with higher employment rates.

In the functioning democracies (D-Index above 8) the correlation is particularly strong. There are some challenged countries (D-Index between 5 and 3) that have very low unemployment (Thailand, Cambodia, Nepal, Belarus). We have not dug into why, but expect it is a unique analysis for each of those regions.

Thus far we have seen a good correlation between personal sovereignty and employment, average wealth, and income inequality.

Of course, there is a balance. If you become too decentralized, these advantages will likely reverse. If there is no universal rule of law, no widely accepted legal framework, and no social services to support the bottom of Maslow’s hierarchy, things will break down. However, to the extremes that modern civilization has evolved, we do not yet see that reversal point.

Competitiveness and Decentralization

If more decentralized societies are better, is it also true that more decentralized marketplaces and industries are as well. We can roughly map types of competition to our axis as follows:

  • Pure competition: Standardized product delivered by many suppliers.
  • Monopolistic Competition: Similar products delivered by many suppliers, who differentiate on features and set prices appropriately.
  • Oligopoly: Few suppliers and difficult for new suppliers to enter.
  • Monopoly: Single supplier which can control market prices.

We use the World Economic Forums Competitive Index (WEF-CI) which ranks countries by how they are structured for competition:

We define competitiveness as the set of institutions, policies, and factors that determine the level of productivity of an economy, which in turn sets the level of prosperity that the country can achieve

Using the same D-Index score for countries, we can map this WEF-CI as shown below. A higher WEF-CI score, the more competitive industry is within that country. We can see that there is a high correlation between the two (R² > 0.5), which may well be bi-causal. Better democracies put in place better frameworks for competition, with antitrust regulations and authorities. Likewise more competitive environments may lead to the development of more democratized institutions.

We may have overfit the curve here, but it is interesting to see competitiveness plateau for Hybrid Regimes. Perhaps there are some fundamental social structures that need to be in place for competitiveness to grow, so the difference between having a D-Index of 3 and a D-Index of 7 is not meaningful in this regard.

While the correlation between D-Index and wealth / wealth equality, combined with the above graph would seem to imply that competitive environments also drive wealth and inequality, the direct correlation between the two is actually very weak. Some of the reasons may be outlined here.

Happiness and Decentralization

Perhaps more important for well being than employment and wealth is happiness. As parts of the world move towards late stage capitalism, there are many who live very good lives on relatively less. The World Happiness Report is published by the United Nations, and we see it mapped against the D-Index below.

This is a stronger correlation that we anticipated. Personal sovereignty — more control over the decisions in your life — are obviously very well correlated to happiness. Again we point out that in the upper left quadrant the correlation is particularly strong, and that the trend line is not maxed out. We can expect that more decentralization will lead to even higher happiness.

Holistic Social Progress

Finally, we look at the Social Progress Index from the Social Progress Imperative which measures fifty-four variables around well being. The index does not use economic indicators, and focuses on outcomes rather than inputs. Thus, it may be the best measure of well being that we consider here.

As you can see above, this is the strongest correlation with democracy / decentralization. By ignoring economic indicators this is particularly telling — when individuals have more control the outcomes are better.

Summary

Of course, we have barely scratched the surface here, and need to look for more data sets and counter arguments.

However, at the highest level, there are no surprises. More personal autonomy within systems, governments and nations that promote and support personal authority leads to better individual lives, across almost every measure.

This begs the question. Can we do even better? Until recently the answer was “maybe, by a little bit.” But recent advances in technology that allow for better management of trust, responsibility and identity within decentralized structures imply that we can do not just a little bit better, but significantly better. This is not just within the frameworks of nation states, but also within other structures in which we live our lives: capitalism, digital experiences, social structures. Any idea can be taken too far, but it is reasonable for us to explore if further decentralization leads to further increases in individual well being.

This gives support to the active experimentation now being done with block chains, smart contracts, and other decentralizing technologies. Instead of being skeptical of this trend, we should actively engage it.

Addendum: Other Interesting Correlations

The point of this post was to look at the trend lines, not individual points…but if you want to see which country is which, click on the version of this post here.

There is a lot of ‘by country data’ that we can plot against the D-Index. Here are few more for fun.

Murder rate per 100,000 shows that being a full democracy has benefits. As soon as you move into Flawed Democracies (6 < D-Index < 8) there is an obvious change in pattern. Why this one group of countries should be so high (other than the USA’s lax gun laws) is not obvious.

The number of people who say religion is important (0 is unimportant; 1 is highly important) is nicely correlated to more oppressive nations. Trying to tease out correlation versus causation on this one is well beyond the scope of this note.

Weeks of Maternity leave seems like it should correlate well, and it does, although not to the extent you might expect. Many challenged countries do better than the USA on this variable, for example. The USA is a clear outlier, with no guaranteed leave (it is the point at D-Index 7.9, leave of 0).

There is not too much data on Leisure hours, but what there is correlates relatively well.

If there are other data sets which would be meaningful in this context, let me know.

--

--