Monday, January 24, 2011

Moderate Tax Sweden

The economic success of Sweden has been one of the most powerful arguments of the left. If taxes are bad for the economy, why does the country with the highest tax rates in the world have an widely admired standard of living?

My argument, that Scandinavians due to their culture (work ethics, cooperativeness, trustworthiness, civic-mindedness etc.) are more productive than other westerners, doesn't work in Sweden. The reason is that Swedes are like fish in the water with regards to their norms and culture. They are generally not aware that they and other Scandinavians are much better behaved than other westerners. Instead the public and the elites naively attribute Sweden's better functioning society to the Social-Democratic welfare state.

This makes it difficult to argue against the welfare state for true believers. As for foreign liberals, the challenge is even harder, since outsiders can only remember a few facts about any small society (Sweden=blondes=well organized=affluent=welfare state).

So how did free-marketers ultimately convince the Swedish public to reform the welfare state? The answer lies in the details of modern Swedish economic history, something which Swedes are aware of but outsiders often are not.

You see, Sweden had very high rates of growth when the welfare state was only slightly larger than other developed countries. The period 1870-1970 is sometimes mentioned, when Sweden had the second highest growth rate in the world, below only Japan.

For simplicity, let's focus on the post-war period. At that time, the Swedish public sector was only slightly bigger than other developed European countries. And for the next two decades, Sweden's Golden Age, the expansion of the welfare state was no faster than average. Around 1960, Sweden was not an outlier in terms of economic policy. The welfare state only exploded in the late 1960s when the left was ideologically radicalized and the right marginalized.

What happened consequently turned out to be the doom of Social Democracy. The rate of growth slowed. This is not only relative to poor European countries that were catching up after the war, but also countries on the technology frontier such as the United States. Sweden lost ground in terms of living standards, while exports slowed and employment stagnated.

The expansion of the welfare state stopped by 1990, and slowly the country began to reform. There was economic and political chaos because of the (largely coincidental) 1991-1993 economic crisis, which makes it hard to tell a consistent story for those years. However it is undeniable that the Swedish state has been in retreat from 2000 onward, not the least thanks to Fredrik Reinfeldt and Ander Borg.

Coincidentally or not, economic performance is back up.

Sweden is becoming a moderate tax country again. This has made my life much easier when debating taxes and economic performance with leftists who refuse to acknowledge the innate superiority of Swedish work ethics. Soon, it will be very hard to paint Sweden as a Socialist Utopia. This is not because it is no longer Utopian, but because the country is no longer that socialist (at least in economic policy). This development also means my personal taxes have gone down a lot, although to be honest as a tax-nerd I care much more about the implications for the policy-debate.

Thursday, January 20, 2011

Breaking Matt Yglesia's Heart

Matt Yglesias is one of the smartest left-of-center pundits. He is rare because he is both knowledgeable about social science and capable of applying it to public policy. Informing policy is the main purpose of social science, so it's good to have a few guys around who can act as links between scientists and pundits.

Yglesias is clearly fascinated by the Nordic countries. He belongs to the majority in the American (and Scandinavian) left who believe that favorable social outcomes in Nordic countries are caused by welfare state policies. If America copied Nordic policies, America would achieve roughly the same outcomes.

I belong to the people who believe that Scandinavia, like Japan, is a cultural outlier. Scandinavians score unbelievably high on things such as work ethic, trustworthiness and cooperation. This interpretation of causality is different: Scandinavians has a large welfare state because they are homogeneous and culturally unique. When other countries (say, southern Europe) copy Nordic policies, the results tend to be worse, because the population is different.

Perhaps the most important social outcome is educational attainment. The fact that the United States has a large achievement gap in education is the root cause of many problems and injustices (the only comparable inequality is crime, but crime is probably to a large extent due to the lack of human capital). In comparison, lack of material wealth is not a major problem in the United States, as African-Americans and Hispanics earn about the same income as the average in Western Europe.

If liberals are right about the fact that America could eradicate the ethnic education gap through copying Scandinavian policies, it would be a disgrace if we didn't do it immediately.

However, empirical evidence suggests that liberals may be wrong about this interpretation of causality. The reason I can claim this with confidence is because I know that the achievement gap in Scandinavia is roughly the same as in the United States. Until recently ethnic minorities were very few in the Nordic countries, so the achievement gap didn't influence the mean score, and was not apparent to outsiders. But not anymore.

During the last 3 decades, the foreign born population in Sweden, Norway and Denmark (but not so in Finland and Iceland) has grown exponentially. First and second generation immigrants from outside of Europe are now over 10% of Sweden's population.

It turns out that the Nordic school system and society at large is no better at closing the achievement gap than America, or at least not yet. One reason may be time. Contrary to conventional wisdom among liberals, the United States has been trying hard to close the achievement gap since at least the 1960s, with a lot of resource and some success. Scandinavia is just starting to become aware of this issue.

Let's also repeat the obvious: The American public school system is simply superior to the Scandinavian system, because America is richer and can afford to spend much more on public education.

Before presenting you with statistical evidence, let me cite Scandinavian sources; if you don't trust me perhaps you will trust them:

DN (largest daily newspaper in Sweden), 2010:
"Only 25 percent of students born in Somalia finish primary education with grades sufficient to qualify them for entering college, compared with 82 percent of students born in Finland. [and 91 percent of native Swedes]"

Swedish National Agency for Higher Education 2009:
"Few students with African origins continue to college"

Skolvärlden (Swedish Teacher Union Journal) 2010:
"Children who have immigrated to Sweden leave the ninth grade with significantly worse scores than children born in Sweden. This is regardless of whether they have received throughout their primary education in Sweden or not."

OECD, The Labour Market Integration of Immigrants in Denmark, 2007:

"the gap in educational attainment between the second generation and persons of Danish origin has grown substantially over the past decade....there is a divergent trend, similar to what has been observed in Germany"

"the second generation in Denmark has an about 20% lower chance of completing a qualifying education than Danes without a migration background."

"about twice as many students from the second generation drop out of upper secondary education as do students of Danish origin"

Statistics Norway, Immigration and immigrants 2008:
"Large differences [between immigrants and natives] in education levels"

"Source Country Differences in Test Score Gaps: Evidence from Denmark"

"even adjusted [for socioeconomic characteristics] gaps [in PISA between native Danes and first and second generation immigrants] remain sizeable and statistically significant. The adjusted gap is largest for students from Lebanon (0.80 SD), somewhat smaller for students from Turkey and former Yugoslavia and Afghanistan (about 0.55 SD), and smallest for Pakistan (about 0.25 SD). Thus, while the results indicate that a major portion of the test score gaps can be attributed to less favorable socioeconomic background factors of immigrant students, educational achievement of young immigrants is poorer than that of native Danish youngsters even after conditioning on the socioeconomic status of their family."

Most of the immigrant population with social problems in Scandinavia is from the Middle East and Southeastern Europe. These groups have few problems in the United States.

However when I pointed this out, many readers remained unconvinced. They want me to look at African-immigrants in Europe, the group considered most comparable to African-Americans. When data is available I will present the figure for all Africa, and when it is not for whatever large group that is available. There are few Mexican immigrants in any Scandinavian country so while a good comparison group for Hispanics is hard to find, I will show you some figures for Chileans.

I will attempt as much as possible to look at children who were born or grew up in Scandinavia, and not recent immigrants. I will look at both Sweden and Norway. From what I understand the situation is even worse in Denmark.

Let's start with the biggest problem in America, the ethnic achievement gap. For High-School, I use Heckman's figures rather than Census data, because there is reason to believe that Census figures under-estimate the gap due to the GED. For college I use Census data.

This we already knew. But how about Scandinavia? Not a pretty picture.

There are some exceptions, such as the high-School enrollment rate of Ethiopians/Eritreans in Norway. However the overall results cannot be denied: There is a massive Majority/Minority achievement gap in Scandinavia, comparable in size with the gap in the United States.

Keep in mind, Scandinavia is doing so poorly despite the fact that the minority population is small, no more than 5-10% of the student population. In the United States, the figure is closer to 40% of pupils.

It would be interesting to see how the left explains this pattern, given that they believe that the Nordic social system (rather than Scandinavian norms and culture) is superior to the American one. If that's the case, why are those without Scandinavian culture doing so poorly under the exact same system? And why are Scandinavians doing so well in the United States, under another system?

My contrarian conclusion is that the Nordic countries should be learning from the American public school system and how it deals with disadvantaged students. In particular, it is important to spend much more, put a lot of emphasis in teacher education, and build external support structures outside the classroom. Another important task is to make sure immigrants children speak the language as well as possible. The politically correct emphases on home-country-language instruction in Swedish schools may be harmful.

One important thing America can learn from Scandinavia is daycare, as James Heckman has argued.

One reason immigrants are doing poorly in Sweden and Denmark is ironically that their parents don't take advantage of day-care, even though it is virtually entirely tax-financed. One reason is that daycare is linked to having a job, which is a terrible idea (children of parents without jobs need daycare most). Another reason is that Muslim immigrants don't want their kids to adapt Swedish values. Perhaps in time we should move to make daycare mandatory in Sweden, with exceptions where it is needed.

Friday, January 14, 2011

Education spending in U.S states and Test-Scores

Some readers wanted me to adjust for the cost of living when comparing education spending and test-scores. To my knowledge no perfect state by state adjustment exists, so I used this one produced by MERIC. The figures seemed plausible (D.C high, Tennessee low).

What I am doing here is comparing education spending per pupil with test scores of 8-graders in different American states, as measured by NAEP (which is considered a reliable comparison of pupil knowledge in different states). I use the average of math and reading.

I repeat that correlation does not equal causality. It may for example be that states who spend the most have the best educated parents. However, establishing statistically significant correlation is a necessary first step in arguing for causality, especially when we don't have controlled experiments.

First, let's start with where the debate between liberals and libertarians is currently: the correlation of state spending and test scores.

At least the correlation is clearly positive, unlike last time when I included other nations as well as U.S states. But the correlation isn't statistically significant, and no self-respecting Libertarians would let liberals get away with using a non-significant relationship in a debate.

Now let us do two simple adjustments. First, I divide the students into three demographic groups. The first group is The Majority, the weighted average of non-Hispanic White and Asians. I add those because the sample size of Asians are too small in many states (Whites alone produce roughly the same results).

Whites and Asians are 60% of the U.S student population. The second sample is Hispanics, 22% of students, followed by African-Americans who constitute 17% of the U.S student population.

First, White and Asians:

Similar results for Hispanics:

Lastly Black pupils:

Notice again the parsimonious power of the crude demographic and cost adjustments: the correlations are about twice as strong, and the relationship is now statistically significant for Whites, Asians and Hispanics.

Admittedly the correlation between spending and test-scores is still weak for African-Americans. I suspect the reason is selection: that the best scoring African-Americans tend to live in rural states with few minorities. (The seven highest are Montana, Vermont, North and South Dakota, New Hampshire, Wyoming and Hawaii). These states spend less than the national average. However the African American parents who move to Hawaii are probably different from the ones who remain in Detroit or the South. Also, states with a large disadvantaged population have their resources spread thin compared to Montana and Wyoming.

So I removed all states where the Black student population is small. This is not strictly kosher, especially since I did it after observing the results. In my defense I did a OLS-regression of the full sample and weighted by state population, and the relationship between spending and test-scores is the same for blacks and whites.

Even though they are passionate about increasing education spending, liberals have not used these simple and effective arguments against anti-education spending libertarians.


Let me put the top 10 and bottom 10 states for African Americans. The Star* signifies a Red-State (a state that was won by Bush in 2004):


1* Montana
2 Vermont
3* North Dakota
4 New Hampshire
5* South Dakota
6* Wyoming
7 Hawaii
8 Massachusetts
9* Texas
10 Delaware


42* Nevada
43* Nebraska
44 Rhode Island
45 California
46 Wisconsin
47* Mississippi
48 District of Columbia
49* Alabama
50* Arkansas
51 Michigan

Before I did the comparison, I assumed that since Blue-States are richer and probably more tolerant towards African-Americans, they would have higher scores. But it turns out the opposite is true, the population weighted NAEP score of African-Americans in the 2004-Republican states is 2 points higher than Democrat states.

This is despite the fact that Blue-States spend significantly more than Red-states on education, both before and after controlling for the cost of living. Yes, the Deep South still has low scores, but this is compensated by above average performance in Texas, the Mountain States and the Midwest.

Ironically, the Blue-States are equivalent to the United States in the PISA debate (more resources, worse outcomes) and the Red-States the equivalent of Scandinavia or North-East Asia (less resource, better outcomes). If we were as prone to jumping to conclusions that support our ideological biases as the left or libertarians, the debate would be over. This "proves" once for all that lower-spending on education causes better outcomes for minorities!

However I suspect the reason for this contradiction has to do with insufficient demographic adjustment, just like the PISA-debate. Blue-state are more urban, so that their schools have to deal with more disadvantaged students. The schools are not necessarily worse, or at least this comparison cannot tell us if they are. We need much more careful studies to determine this, controlled experiments or at least a before-and-after methodology that takes into account the background characteristics of the students. Just reporting average scores is not enough.

Liberals are rarely willing to cut the United States slack for having a different population mix than Scandinavia. In the education debate, the are getting a dose of their own causality-ignoring medicine from libertarians.

Friday, January 7, 2011

California vs. Texas

Paul Krugman is now claiming Texas doesn’t have a well managed economy or something. To be frank, I didn’t read his column, because I have grown tired of the man and his dogmatic and predictable views. He has been caught cheating so many times that I now trust Krugman's figures as much as I trust Glen Beck's.

But at least I have an excuse to write an easy post comparing California and Texas. A lot of people have done this, so here are my two cents.

I use budget data from the Census Annual Survey of State Government Finances and job and per capita income data from the Bureau of Economic Analysis.

Last year, high-tax California had a population about 50% larger than Texas, a deficit 220% higher, and a debt 380% higher. It's safe to say that low-tax Texas is in a better fiscal shape.

Since 1970, Texas has added 60 percentage points more jobs than California. A lot of this is due to faster population growth. But not entirely. In 1970, the employment to population ratio was 1% higher in California, while in 2009 it was 6% higher in Texas.

In 1950 the Golden State had 40% higher per capita income than Texas. In 1970, the advantage was still over 30%. By 2009, the difference had shrunk to only 10%, without taking into account the higher cost of living in California.

(this is how to read the graph: Texas is about twice of what Texas was in 1970, and California 60% higher than California in 1970. The graph shows that Texas has grown faster since 1970, not that it has surpassed California in terms of levels).

It appears that Texas is doing better than California not only fiscally, but also in terms of aggregate job and income growth.

One thing I would warn about is exaggerating California’s debt problem. It’s true that they have mismanaged their finances, and expanded government beyond what they can afford. However California is still extremely wealthy, with a total GDP of about 1900 billion dollars in 2009. This is about the size of the entire Italian economy, and still larger than Brazil, India, or Russia.

So while their tax base may appear narrow, their entire economic base is very wide. The debt to GDP ratio for California alone is still below 10% (or 80%, if you add the national debt).

Also, let's not make too strong policy-inference from the short-term mortgage-bubble that is currently distressing California. Policy should be based on evaluations of long term performance. I argue above that the long term trend also favors Texas.

Monday, January 3, 2011

The relationship between education spending and test scores

Good public policy is impossible without understanding causality. The most common logical mistakes in the policy debate is the tendency of confusing correlation with causality. One common reason correlation is confused with causality is the failure to make demographic adjustments.

A demographic adjustment without some form of exogenous variation can rarely tell us what the definitive truth is. But it can often be used to discard weak arguments. Usually, I focus on criticizing the left in this regard, but let's make an effort to be intellectually honest and scrutinize the right for once.

The argument of the right is that educational spending is wasteful, and doesn't produce outcomes. They base this on the fact that the United States already spends more on education than almost all other nations, and more than ever in history.

John Stossel for instance writes that:

"If money were the solution, the problem would already be solved"

The difficulty is that education spending goes up when social needs go up. This is the classical head-ache pill example: just because people with worse head-aches take more pills, it doesn't mean the pills are ineffective or even causing the pain. The same argument is also true of spending on the police. Cities with more crime have more police, not because the police are causing higher crime rates, but because they are a response to high crime.

So for instance the entity that spends the most on education and has the worse test scores in the U.S is the District of Columbia. This doesn't prove anything, because D.C also has much more poverty and many more disadvantaged students. The relevant policy question is what would happen to test scores if D.C spent less.

Let's start with a simple cross-country regression of the normal, unadjusted PISA-scores (the ones the media keeps reporting) and education spending per pupil, as measured by the OECD.

It turns out that the correlation is negative (-0.2). The United States spends the second most of any country, and has below average test-scores. Ethnically homogeneous Japan, South Korea and Finland spend at average rates and have the best test scores. Tiny, ethnically homogeneous and "hungry" Estonia spends less than half as much as the United States and Norway on education but has far better test scores.

Argument over. Doing the comparison in the New York Times way, Conservatives win, and Liberals lose.

But what if we do it my way instead? Let's do a crude demographic adjustment, and compare European-origin groups to other European-origin groups, just as I did before. The methodology is explained and defended in detail here.

For Europe, I remove first and second generation immigrants. For U.S, I remove Latinos, Asians and all other groups than non-Hispanic whites (two thirds of the U.S population is non-Hispanic white).

Now, prepare to witness the power of the simple demographic transformation.

The relationship between spending and outcomes is now reversed. More spending is indeed associated with better PISA scores. The United States does well, as predicted by how much she spends on education. Finland with her admired education system is still an outlier, which I guess indicates there really is something special about Finland and their school system.

The graph suddenly makes more intuitive sense, doesn't it?

But we have more variation in data we have not used. Why not break out each U.S state and treat it as a separate observation? We have comparable data on 8-grader test scores for each state from the NAEP which I have converted to PISA-equivalents (method explained here).

We also have access to education spending per pupil for each U.S state, courtesy of U.S Census.

Again first we do the standard, unadjusted PISA-scores for this pooled sample of countries and U.S states.

A flat, slightly downward sloping curve, countries and states that spend more don't appear to score any higher.

Notice that Utah, Finland and South Korea outscore California, even though they spend half as much (they don't need to spend more, because they lack California socially disadvantaged masses). Minnesota beats Sweden, but generally U.S states do poorly, even though they spend more than Europe.

Another victory for Republicans. Clearly doubling money on education has no effect, and may even - somehow - make your countries children score a little less.

Now, look at the exact same graph, only with my standard crude demographic adjustment (For Europe, first and second generation immigrants are removed. For U.S states, all demographic groups other than non-Hispanic whites are removed from the sample).

Again the results are reversed after a demographic adjustment. The advantage of the U.S and the advantage of rich U.S states that spend a lot on education suddenly becomes apparent. More public spending on schools is associated with better outcomes. The mystery of the missing correlation is solved.

Lastly, let's look at test scored of African Americans and state educational spending.

Of course these graphs alone do not prove that the driving factor is education spending itself. In theory, it could be because students in rich states are different, for example in terms of family income. But the relationship still represents powerful suggestive evidence in favor of public expenditure on education.

However the left in the United States doesn't use this argument, because they are ideologically averse to demographic adjustments having to do with race and ethnicity (most of them consider all statistical generalizations about race and ethnicity somehow offensive, regardless of why you are doing it).

The result of liberals political correctness is that they are depriving themselves of a very important argument in a very important debate.

Sunday, January 2, 2011

Nicholas Kristof makes a fool of the New York Times

Nicholas Kristof has given a raving review of the book "The Spirit Level" in the NYT. He is clearly unaware of the debate in Europe, where the "Spirit Level" has largely been debunked due to poor scientific quality.

I have written about this, and would like to remind you of 3 facts about the book "The Spirit Level" (links at the end):

1. The book is scientifically dishonest by never telling the readers that it represents a minority opinion within social sciences. Higher ranked researchers in better schools using more rigorous methods have come to the opposite conclusion about income inequality causing lower life expectancy. Yet the reader of the book, or readers of the New York Times, would never even know the other research existed.

Wilkinson and Picket give their trusting readers the impression they are reading novel research results. In fact, they are reading claims that were debunked years ago.

Let me refer to scientific authority, because that's what Kristof does. Princeton Professor Angus Deaton is currently the 57th highest ranked economist in the world. He is an expert in health economics, statistical methods and one of the foremost expert in determining causality. He is in often mentioned as in the running's for the Nobel prize. Deaton is a typical center-left academic, not some right-winger.

In 2004, Angus Deaton reviewed the research on income inequality and life expectancy for the Journal of Economic Literature, the most authorities such journal in economics. Deaton in particular looked at the empirical evidence for the claims of Wilkinson. He concluded:

"it is not true that income inequality itself is a major determinant of public health. There is no robust relationship between life expectancy and income inequality among the rich countries, and the correlation across the states and cities of the United States is almost certainly the result of something that is correlated with income inequality, but is not income inequality itself"."

Kristof repeats the claim in the New York Times that income inequality causes poor health outcomes in American states and cities. This particular hypothesis was looked at by Deaton and Lubotsky in 2003. They analyzed the link between income inequality and mortality, controlling for demographic differences. They found (from abstract):

"Conditional on the fraction black, neither city nor state mortality rates are correlated with income inequality"

2. Those nice convincing graphs in Spirit Level graphs are to a large extent based on manipulating and cherry-picking.

For instance, I, some blogger in his pajamas, forced Wilkinson to admit publicly in the Wall Street Journal that he had established no statistically significant relationship between life expectancy and income inequality across countries.

Most of their other claims could not be replicated either (although I didn't test everything). For instance, Wilkinson and Picket assert that they have proven that income inequality cases lower innovativeness. I was suspicious of their claim that the United States was one of the least innovative countries in the world, at the same technological level as Portugal.

It turns out the "distinguished" scientists who convinced Kristof by citing "mountains of data to support their argument" picked the innovation data from amateur internet site Nationmaster! Their claims was not supported using patent data from the World Intellectual Property Organization.

3. Wilkinson and Picket confuse correlation with causality.

Sure, they pay lip-service to causality. But they never tell the readers that among all the papers they cite, there is not one single article that has shown any causal effect of income inequality on health or social problems.

Every single article Wilkinson and Picket cite (and they are lots of them), is some version of a correlation studies, sometimes with few and sometimes with more controls, but never with identified causal links.

You don't believe me? It's true.

Wilkinson and Picket rely on 2 meta-studies to create the impression of massive external scientific support for their view. The big one, with close to 200 papers, was written by no other than themselves (!). You can read it here:

In the paper Wilkinson and Picket explain their criteria for determining if a paper is evidence for their hypothesis (inequality causes poor health).

“We classified them as wholly supportive if they reported only statistically significant associations between greater income inequality and poorer population health”

Did you catch the key word? "association". In terms of causality, "association" is the equivalent of correlation (although association doesn't have to be linear). Association is what researchers write when they have failed to establish causality.

This is how Wikipedia explains the term:

“In quantitative research, the term "association" is often used to emphasize that a relationship being discussed is not necessarily causal”

Wilkinson and Picket gather a bunch of articles (87 have a statistically significant relationship) where there is some form of correlation between inequality and health. But of course, poor health can cause inequality, and third factors (say, drug use or unemployment) can cause both inequality and poor health. Establishing causality here is not trivial, such as from smoking and mortality, its central to the debate.

In fact, it is hard to think about any social problem that doesn't depress income for those afflicted (and thus create income inequality) as a side effect. So typically what will happen is that there is a weak and not very robust correlation between inequality and poor social and health outcomes, which vanishes when researchers do a more careful analysis with more controls.

So proving a causal link between inequality and health/social outcomes is not a detail, it's essential to accepting their story.

But Wilkinson and Picket never establish causality. They never even claim to in the scientific version of their paper. There, facing a less gullible audience, they admit that they are not showing causal effects, just statistically significant "associations".

But in the book, and on the internet, and to their audiences, and to trusting suckers such as Nicholas Kristof, Wilkinson and Picket give the dishonest impression that the papers they are citing contain causal evidence.

Wilkinson and Picket are relying on the lack of scientific sophistication among the general public. The public doesn't always understand the difference between correlation and causality, and can obviously not be aware of the studies in economics that already debunked the claims in the Spirit Level. But isn't it the job of the New York Times to ensure scientific standards for their readers?

Whenever pushed into a corner, Wilkinson and Picket rely on the number of citations in their book. However the number of studies happens to be irrelevant when the problem is reverse causality. 20 or 200 or 2000 correlation studies of the same spurious relationship will give the same answer. If you have failed to identify a causal link, all the 200 correlations are equality worthless.

The New York Times is allowing Nicholas Kristof to embarrass their newspaper. Don't simply trust everything Wilkinson and Picket claim because you dislike inequality (so do I) and are easily impressed by academics (I used to, but than I grew up). Academics sometimes fall to the temptation to abuse their authority and our trust to package their opinion as science or to sell books. Do your homework.

From the start I noticed that these two are dishonest people, which I don't like. I have thus gone after them aggressively. Since the book is such a house of cards, it was not hard.

Saturday, January 1, 2011

How well do above average American States do compared to above average countries in test scores?

Over at National Review Online, Reihan Salam has some complaints about my PISA-comparison.

First, he notes that even if European-Americans and Asian-Americans are doing well, minorities are not, and they are going to be a larger and larger part of the U.S population. This is completely true.

However my analysis tells us two new things:

* The problem is not likely to be the American public school system itself. If it was the education system that constituted the problem, Asian- and European-descended students (70% of the U.S population) would not be performing so well relative to other nations. After all, they are in the same government controlled system. Studies that use more sophisticated methods to control for school quality come to the same conclusion: It's not the schools.

* Europe has a similar achievement test gap between majority/minority. It's just that the size of the European minority population is still small, so it doesn't affect the national average as much.

But if your aim is to isolate the treatment effect of a policy (in this case the design of the school system), the size of the group impacted is irrelevant. Turks in Sweden have approximately the same gap with Swedish natives as Mexicans in California to California natives. This should conclusively end one discussion: the Scandinavian school system has not discovered how to solve the majority/minority achievement gap.

If you want to explain why Sweden has less aggregate economic and social inequality, the relative size of the minority groups in question matter. If you want to compare the efficiency of school systems, it doesn't. It is immaterial for what we are doing, a comparison of how well the school system closes the achievement gap, that Turks are less than 1% of the Swedish population and Hispanics 30% of the Californian population. It is sufficient to show that they have the same gap. If anything, European schools should be doing better considering the smaller size of the disadvantaged groups.

Jumping up and down in excitement, and congratulating Swedish and other Social Democratic school systems for having "solved" the achievement gap while sending fact finding teams there to learn from their system is ridiculous. Their system hasn't solved anything, the same gap is there, they just have few minorities so the gap influences national averages less and is less apparent to outsiders.

So under closer scrutiny it turns out that the "secret" of Scandinavia's lack of a achievement gap is that until recently, they didn't let many minorities into their countries. Once they started to, the achievement gap among socially disadvantaged groups emerged, a little smaller in magnitude but likely due to the same causes.

In the 1960s, a naive liberal in the U.S could be excused for assuming that Sweden's success was due to their education or welfare policy, and that Sweden had "solved" societal problems such as ethnic achievement gaps. Other countries could just copy whatever Scandinavia was doing in order to deal with their social problems.

Now that we have all seen footage from Scandinavian ghettos, as well as national statistics from Sweden and the rest of Europe confirming an achievement gap with minorities, it should be clear that this theory was wrong. So why are we still spitting on American public schools and congratulating Social Democratic ones for producing roughly the same outcomes?

2. Salam believes that the U.S should be doing even better, and asks about the performance of Greenwich (Connecticut).

"The NEA, and Tino, would have you believe that it is very, very impressive that Greenwich is treading water with Yazoo...."the United States, in terms of GDP per capita, is closer to being the Greenwich [Connecticut] of the OECD than the Yazoo City [Mississippi]."

He has two points here. First, he believes that with America's resource advantage, the U.S should be outperforming other countries even more. I might write more on this in detail later.

The resource gap of the U.S vs. Western Europe is not the same as that between Connecticut and Mississippi. As I have written here, Connecticut has 110% higher per capita income than Mississippi. Meanwhile the United States has "only" 36% higher per capita income than Western Europe.

If we instead of Mississippi compare Connecticut with the U.S average, it fits better. Connecticut is 35% above the U.S national average in per capita income, and the U.S national average is about 35% above Europe. In this sense the United States is equivalent of a (albeit much larger) 'Connecticut of the OECD', which is what Salam wrote.

Nationally, Connecticut as a whole scores a little less than 0.2 standard deviations above the U.S mean. Well, my analysis suggested Americans of European-origin score a little less than 0.2 standard deviations above Western Europeans in Western Europe.

So if this is the criticism, it turns out that the U.S educational advantage is approximately what the U.S economic advantage would predict.

I can't fault Mr. Salam for wanting more output for the input American tax payers are spending. However my analysis suggests that unlike conventional wisdom on the political right, public education spending is not just wasted. It appears successful in buying America better test scores, for each given demographic group, and of course with diminishing marginal returns.

Second, Salam is not sufficiently "impressed" by the average American advantage. Let's see if we can impress him.

The data I presented was national averages. In the case of the U.S, this was the average of rich cities such as Greenwich Connecticut and poor cities such as Yazoo Mississippi. When you aggregate a large group (200 million people) the variance goes down, as Greenwich and Yazoo cancel each other out.

This is important to keep in mind when comparing with small outliers such as Hong-Kong and Finland. Even if the U.S beats Europe on average, the best performing tiny part of Europe will likely beat the average of the U.S. What I was impressed by was the fact that one large group (all Americans from Europe) beat another large group (all Western European native born) by a sizable margin, as well as beating most of the small, top performing outliers.

But I would be happy to disaggregate to defend my argument.

PISA doesn't report the score of U.S states, because the sample sizes are too small. However there is another similar international test of math and science of 8-graders that does have data on individual states, TIMSS. They also have data on 2 U.S states, Minnesota and Massachusetts (both are similar to Connecticut in terms of test scores nationally, according to other data).

Remember that the United States scores relatively worse on TIMMS than PISA, especially compared to Asian countries, because PISA includes reading (which Americans are good at), whereas TIMSS is only a test of math and science.

Massachusetts and Minnesota each score higher than every single European country, although Finland is not included.

The achievement gap between Minnesota and Sweden is as large as the gap between Sweden and Bulgaria!

Keep in mind that this is the average of Minnesota and Massachusetts, including immigrants and minority students. If you were for example to break out only Asian-Americans in one of those states, the results are going to be stratospheric.

For fun, I also used a third large test of 8-graders (NAEP) which allows me to break out Asian-Americans in Connecticut, but requires more work to make it comparable with PISA. If Asian-Americans in Connecticut score as high relative to other Americans on PISA as they do on the NAEP relative to other Americans, they would beat every single Asian country in PISA. So would Asian-Americans in Texas and New York and several other states.

Note: this second graph relies on translating scores on one standardized test of 8-graders (NAEP) to PISA-equivalent scores, assuming that your group scored 10% (in standard deviation terms) above the American national average on the NAEP means your group would also score 10% above the American national average on PISA. I know that those readers who are resistant to my argument are going to nit-pick and are going to to be reluctant to accept the comparison.

However regardless if you buy my second graph, what is very important for readers to understand is that the first graph with TIMSS is not based on me doing any sort of adjustment of the data. It's simply the average of the scores, as reported by the Department of Education, on a test that is internationally standardized.

This doesn't definitively settle the debate about the quality of education system, because of biased selection of Asian immigrants. But I hope this example illustrates that affluent groups in the United States destroy the international competition, and are competitive even when compared to small rich countries such as Hong-Kong.
Google Analytics Alternative