New article on Sweden in the Milken Review

My latest piece on Sweden is out now in the Milken Review. This time I try to say something about how and why Sweden became the country it is today. Previously, I have focused on what happened, when did it happened and what the consequences were. Saying something about why it happened is much more difficult. In short, it is easy to paint a picture of wise intentional planning in retrospect, but that is typically not true when you look closer at what really happened. Serendipity and unintended consequences played important roles.
From the introduction:
The winding road Sweden has taken has made it difficult to say whether being more like Sweden involves increasing taxes and government intervention in the economy – or whether it means liberalization, deregulation and welfare-state retrenchment. So, before other countries try too hard to become more like Sweden, it is wise to look back at how Sweden came to be Sweden.
An example, based on the research of Thor Berger:
Historians point to the early introduction of mass public education, with the adoption of the 1842 Elementary School Act. The law, which stipulated that every parish must have at least one school, is often mentioned by contemporary politicians as a shining example of Sweden's long commitment to investment in human capital. The policy implication is seemingly clear: political decisions promoted growth early on by mandating public education. That may well be the case. But before jumping to that conclusion it is worth considering the analysis offered by the economic historian Thor Berger of Lund University. [...]
In short, education promoted economic development in Sweden, but democracy at the time did not promote education. Knowing more about what actually happened in Sweden hardly leads to clearer recommendations for other countries.

A Washington Consensus for welfare states?

A new paper (together with Margareta Dackehag and Martin Rode) is now forthcoming in the EJPE: "Are OECD policy recommendations for public sector reform biased against welfare states? Evidence from a new database". The paper introduces a database that summarizes the policy advice in the OECD publication Economic Surveys. We show that the policy advice given in these surveys can be summarized as a kind of 'Washington Consensus' for welfare states, and that the interest in different types of reforms have varied over time:
We quantifiy the perceived reform need and examine its correlates, which led to an interesting discovery: The reform need according to the OECD is highly correlated with the Fraser Institute's Economic freedom index.
We also examine if perceived reform need predicts subsequent policychanges. Using the Comparative Welfare State Entitlement dataset, we find some evidence that a high reform need is followed by lower welfare state entitlements in countries with a right-wing government, and higher entitlements in countries with a left-wing governments - but these results are not causal, ie these changes may well have ocurred regardless of what OECD recommends.

The database is available here.

The naive moldable young?

Can inferior institutions destroy social trust? In a recent working paper (w Richard Öhrvall) we shed some new light on this question by studying social trust among Swedish expatriates in different types of countries. Obviously, Swedish expatriates are strongly self-selected (as we confirm in the paper, mainly on subjective health), so our main interest is in whether length of stay tend to affect trust differently in countries with different types of institutions.
As (perhaps) expected, it does. But the effect is driven entirely by those who were young when they moved to the new country.
The graphs below show how social trust (measured on a 0-10 scale, corrected for individual characteristics) varies with how long the respondent has lived in the new country. Worst, middle and best refer to the destination countries divided into thirds based on their score in the corruption perception index.
What we see is that for those who were older than 30 at the time of arrival, social trust is remarkably stable. Even in the most corrupt destination countries, swedish expats who have lived there several decades are as trusting as those who only just arrived. Among expats that were 30 or less, there is a marked decrease in trust for those who have lived longer in the most corrupt countries. Note however that after 10 years, there is no evidence of a further decline.
(Graphs look similar if we instead use legal quality as measured by the second dimension of the economic freedom index).

The top three countries according to the reputation institute

Here is the top of yet another country ranking:
3. Switzerland — 77.00. Switzerland stands on its own at the heart of Europe despite not being a member of the EU. It is known for its banks and high standard of living, and skiing in the Alps attracts people from all over.
2. Canada — 77.82. Canada is high on the list of countries people would love to immigrate to. With a tolerant culture, a strong economy, and high levels of healthcare it is no wonder.
1. Sweden — 78.34. Sweden has it all: high-quality exports, a tolerant society, low crime, beautiful cities to visit, a high standard of living, a mild climate, and a strong sense of business. It could well top the list for years to come.

Further evidence that trust explains the welfare state (and not the other way round)

In 2011, I published a paper with Christian Bjørnskov, with one of the best titles I have ever come up with:
Bergh, Andreas and Christian Bjørnskov. 2011. "Historical Trust Levels Predict the Current Size of the Welfare State." Kyklos 64(1):1-19.
The ideas is basically that countries with high average social trust (also known as generlized trust or interpersonal trust, and measured by asking people if they agree that mostr people can be trusted) are more prone to creating and successfully maintaining universal welfare states, and that this mechanism is the main reason why trust and welfare state size is positively correlated at the country level (rather than the idea that the welfare state somehow causes trust).
Granted, some might not be entirely convinced by our country level IV-strategy, but there is now individual level evidence published that strengthens our result:
Daniele, Gianmarco, and Benny Geys. 2015 "Interpersonal trust and welfare state support." European Journal of Political Economy 39: 1-12.
Cited from the abstract:
This article argues that citizens' trust in their fellow citizens can play a central role for welfare state support, because it buttresses the belief that others will not use the welfare system inappropriately. Using the fourth wave of the European Social Survey, we confirm a strong positive association between interpersonal trust and welfare state support (controlling for institutional trust). We also show that: i) this link is driven at least in part by the mechanism discussed above; ii) causality runs from interpersonal trust to welfare state support (using a sub-sample of second generation immigrants); and iii) the effect of interpersonal trust appears conditional on the perceived quality of a country's institutions.


Should we use standardized inequality databases such as SWIID?

Here is my implicit point of view regarding the debate between Jenkins (2015) and Solt (2016):
Below is a table (Table 1) from Rudra (2004).
Do you notice anything strange about these Gini-coefficients? Hint: to verify inequality data, I always look at the country I know best, to see if data make sense...

[I will update this post with my thoughts eventually]

Clearly, something is wrong with the data regarding Sweden in the 1970s. The table suggests that inequality in Sweden was at its lowest level in 1975 (at 27.3) and at its highest level just a year later, in 1976 (33.1). In a country like Sweden, inequality never jumps that much from one year to another, and for sure not in 1976. Reexamining the Deininger and Squire database, it turns out that the 1975 value comes from the LIS database, whereas the 1976 value is taken from Statistics Sweden. Most likely, the latter includes capital income and the former does not. Checking other figures reveals that mosty data for Sweden are net household income, but for Brazil gross income is used, and for China the unit is the individual, not the household.

Rudra is not alone. In fact, she is better than many other papers because the inclusion of a table like Table 1 above means that the errors are possible to spot by reading the paper closely. Often, D&S data are just added to the analysis without even a simple visual inspection, which means that the analysis uses incomparable Ginis.

One of the biggest benefits of Solt's Swiid, is that all Ginis are converted to the same typ (LIS-standard), and mistakes like these are avoided.

References:
Jenkins, Stephen P. 2015. "World Income Inequality Databases: An Assessment of Wiid and Swiid." Journal of Economic Inequality 13(4):629–71.
Rudra, N. 2004. "Openness, Welfare Spending, and Inequality in the Developing World." International Studies Quarterly 48(3):683-709. doi: 10.1111/j.0020-8833.2004.00320.x.
Solt, Frederick. 2016. "On the Assessment and Use of Cross-National Income Inequality Datasets." Journal of Economic Inequality (forthcoming).