Overall, the article is excellent, you should read it if the subject interests you. Random comments:
* The article exaggerates the problems by focusing on Rosengård, the worst (major) ghetto in Sweden. Yes, in Rosengård 80% of adult don't work, but for Sweden as a whole the figure for non-western immigrants is about 50%.
* While the first and second generation immigrants are about 25% of the population, many of these are Finns and Scandinavian. The relevant figure - the number of 1st and 2nd generation non-western immigrants - is around 10% based on my calculations from the latest demographic reports.
* The funniest and most revealing part of the article is a Iranian "rapper" quoted as saying "I want to be able to become president [of Sweden]." Sweden is a Monarchy and thus has no President. What this illustrates is that most multiculturalists ideas are directly imported to Sweden from the United States.
Both "rap" and the dream of becoming "President" are from American popular culture. The majority of television programming and movies in Sweden are American. This is not the first time I have heard the "President of Sweden" dream from immigrants.
Immigrant-right activists in Sweden also quote Martin Luther King a lot, which I find quite offensive. You simply *cannot* compare refugees like me who come to Sweden and are granted asylum and receive generous welfare with African Americans who were enslaved, exploited and forced to live in brutal segregation for generations.
The story also shows how isolated some immigrants in the ghettos are from Swedish society. This guy dreams of running for office, and demands that Swedish society elevate him to this position, but doesn't even know the form of government of the country he lives in.
* Because the media seldom reports the numbers and often gives anecdotal evidence to the contrary, most Swedes are not aware that immigration has accelerated the last few years to about 100.000 per year, of which about 2/3 are non-western. We are now taking in as many every year than the record year of 1993 (War in Bosnia), and more as a share of the population as the United States during the 19th century. But if a tree falls in the forest and the Swedish media ignores it, did it really happen? Instead the conventional wisdom claim (lie) is that "it has become much harder to migrate to Sweden".
* Towards the end the article become more misleading, as the New York Times trusts Swedish social scientists. They write "Some experts believe the support for the far right has already reached its limits in Sweden."
First, anyone who believes that the anti-immigrant Swedish Democrats are going to go away either doesn't understand Swedish politics or is engaging in wishful thinking. They are only going to grow, if for no other reason because the immigration problem is not going to go away. While "movement" parties with roots in the early 20th century such as the Social Democrats and Center-party are withering away, energized Sweden Democrats are building a new grass-root movement. Furthermore, now that they are in Parliament, they receive more media attention and are slowly becoming more mainstream.
The "experts" quoted don't seem to realize that the Swedish Democrats have already grown from 5.7% to 8% in some polls since the election no more than six months ago.
Another example is a Swedish political scientists who gives the classic misleading comparison of immigrant sentiment with the 1990s to "prove" Swedes are becoming more pro-immigration. I have written about this method of cheating at lenght here.
The 1990s was a period of severe economic crisis, and also the beginning of the first wave of anti-immigration sentiment (with New Democracy, the first anti-immigration party). As New Democracy imploded due to leadership conflicts and as the economic crisis ended, anti-immigration sentiment declined. It has since gone up again.
The correct comparison would be with 1970 or 1980 or even 2000, not the peak of a frenzy in 1992. That is a little like writing that American anti-Islamists sentiment has been declining, by comparing to September 12th 2001.
And why not just provide the numbers so the NYT readers can make up their own minds? Of those Swedes who offer an opinion, 62% want to reduce immigration. Instead they avoid the subject by writing that most Swedes don't have immigration as their single biggest political concern.
* The immigrants interviewed are indignant and resent Swedish society. They don't accept even the mild criticism of radical Islamism that the Swedish media occasionally lets through. Worse still, they take it personally.
"“It’s hard to watch the news,” he said. “It’s Muslim this, Muslim that. Everything is about how bad we are. The Swedish won’t say anything to your face. But they say things.”"
This guy perceives Swedish television news being anti-Muslim! I worked briefly in Swedish Television when I was in college, the ideological atmosphere was exactly what you would expect. According to a recent survey 87% of journalists in Swedish television are liberal, leftist or socialist. I guess the reason that the guy is upset is that media grudgingly reports terrorist attacks and immigrant riots.
I have long been aware of and concerned by this deep resentment against Swedes and Swedish society among a large number of immigrants. It may be the most under-reported aspect of the problems (no poll has ever been conducted on this). When I was a child, I remember feeling this dark feeling myself. It is a bitter, dangerous sentiment that will never allow you to successfully integrate.
* This article could not possibly be written in a major Swedish Newspaper in the current intellectual atmosphere. The liberal New York Times is way too honest about the problems caused by non-western immigration to Sweden.
You have to admire America as effectively more democratic than conformist Sweden, where a small group of like minded people decide what facts the public can be trusted to handle.
Sunday, February 27, 2011
Sunday, February 20, 2011
Islamists caused overwhelming majority of terrorist deaths in Europe during last decade
This is the blog post that has taken me the longest to write. I went through every single terrorist attack in Europe and North America in the comprehensive RAND Database of Worldwide Terrorism Incidents for the last 10 years (close to 4000 incidents). Terrorists were divided between Muslims or non-Muslims. When RAND does not provide information on likely or confirmed perpetrators and we have no strong suspicions, I assume they were non-Muslim, to err on the side of caution.
Most of Islam is of course non-violent. Surveys show that less than 10-30% of Muslims are openly sympathetic to violence in the name of Islam. I am therefore not writing about Islam, I am writing about Islamism, a minority political sect within Islam.
Looking at all people killed by terrorist attacks in Europe and North America during the last 10 years, 97% was committed by Muslim terrorist, or 4703 of 4873 killed. Most of this is September 11 alone.
Still, even if we exclude the September 11 attacks, the share of casualties due to Muslim terror is 91%.
A lot of the remainder are a number of deadly (and under-reported by western media) Muslim terrorist attacks in Russia. If we just look at Western Europe, the share of terrorist deaths caused by Muslim terrorists during the last 10 years is 79%.
The remaining 68 deaths out of 319 were committed by The IRA and other domestic terrorist. By comparison, the Madrid attacks in 2004 alone killed three times as many people than all attacks by ETA, The IRA, Corsican separatists, right-wing terrorists and all other non-Muslim terrorist attacks in Europe during the last ten years combined.
Remember, I do not include any Islamists terrorist attacks in the Middle East or South Asia or Africa or anywhere else other than Europe and North America. Based on State Department Data and to unimaginable horror these attacks appear to have killed in excess of 10,000 people per year during the last decade.
Unlike Neo-cons, I do not believe that radical Islamism is a threat to our civilization the same way Nazism or Communism ever was. The reason is that militant Islamism is too disorganized.
Lack of organization makes it hard to eradicate militant Islamism, as is no center of power you can knock out to end the war. However it also means Islamists are unable to concentrate the force required to really threaten us. At worst, they can kill a few thousand innocent civilians, which is of course horrible, but hardly on the same civilization-threatening level as Nazis exterminating millions or Communists threatening to eradicate Europe with nukes.
Second, unlike Communism and Fascism, militant Islamism has little attraction as an ideology in the West. During the cold war communist sympathizers infiltrated governments and other key institutions in the United States and Western Europe. Communist American scientists stole U.S military technology and helped Stalin build nuclear weapons. The same will not happen with regards to radical Islam. Today there are very few American scientists who are True Believers (Useful Idiots) and likely to steal nuclear secrets and give them to Bin Laden.
However what does annoy me is when the elite uses statistics to manipulate the public.
In Sweden the public is currently worried about Islamic terror, after two recent incidents with Islamist terrorist. Once again the elites, (media, politician and academics) have ganged up against the public and are trying to downplay the terrorist threat from militant Islam.
You see, while ordinary citizens may have gotten the impression from the nightly news that adherents of militant Islamism are statistically overrepresented in terms of international terrorism, the Enlightened Classes know better. The real reason the poor fools in the public believe there is a link between militant Islam and international terrorism is "islamophobia".
There was a debate in the parliament recently. Both the Social Democrats and the other parties in the left claimed that the main terrorist threat to the Swedish public today is not radical Islam, but right wing extremism...
The Green party representative explained that terrorism "is about emotional and social aspects rather than ideological ones".
This is the left, but the right is not much better.
So how in god's name would you convince the public of the preposterous claim that militant Islam is not the main terrorist threat in Europe? Their method is interesting, both in showing ingenuity in finding ways to trick the public, and stupidity in what they are willing to convince themselves of. The solution is namely to rely on the number of terrorist attacks, rather than on casualties from terrorism!
In Spain and Northern Ireland in particular, there are lots of tiny terrorist attacks by domestic terrorists every year. These attacks typically don't kill anyone, and often don't appear aimed at killing anyone. Characteristic examples from the RAND database:
"The headquarters of the Spanish Socialist Workers' Party (PSOE) in Baranain had several windows broken when it was attacked with stones. The damage is estimated at 50,000 pesetas."
"A group of radicals attacked a Renault dealership by setting a fire which damaged three vans belonging to the firm. A container of flammable liquids and firework rockets were used to start the fire."
Al-Qaida on the other hand specialized on a few spectacular attacks aimed at killing as many as possible.
Thus if we ignore the deaths and treat each attack as equal, you can show that Islamists commit a smaller number of terrorist acts than domestic terrorists. This method absurdly assumes that bringing down the Twin Towers on September 11 is equal to ETA vandalizing some property in Spain (each is one attack after all).
This is for instance what Sweden's largest daily DN does, using a report from Interpol about the number of terrorist incidences in Europe, and concluding that "Islamists terror attacks are unusual in Europe".
My calculations show that DN is misleading it's readers. As usual, when it comes to issues shrouded in political correctness, the public is better off relying on their own impression than "scientists" and "experts" in the media.
Most of Islam is of course non-violent. Surveys show that less than 10-30% of Muslims are openly sympathetic to violence in the name of Islam. I am therefore not writing about Islam, I am writing about Islamism, a minority political sect within Islam.
Looking at all people killed by terrorist attacks in Europe and North America during the last 10 years, 97% was committed by Muslim terrorist, or 4703 of 4873 killed. Most of this is September 11 alone.
Still, even if we exclude the September 11 attacks, the share of casualties due to Muslim terror is 91%.
A lot of the remainder are a number of deadly (and under-reported by western media) Muslim terrorist attacks in Russia. If we just look at Western Europe, the share of terrorist deaths caused by Muslim terrorists during the last 10 years is 79%.
The remaining 68 deaths out of 319 were committed by The IRA and other domestic terrorist. By comparison, the Madrid attacks in 2004 alone killed three times as many people than all attacks by ETA, The IRA, Corsican separatists, right-wing terrorists and all other non-Muslim terrorist attacks in Europe during the last ten years combined.
Remember, I do not include any Islamists terrorist attacks in the Middle East or South Asia or Africa or anywhere else other than Europe and North America. Based on State Department Data and to unimaginable horror these attacks appear to have killed in excess of 10,000 people per year during the last decade.
Unlike Neo-cons, I do not believe that radical Islamism is a threat to our civilization the same way Nazism or Communism ever was. The reason is that militant Islamism is too disorganized.
Lack of organization makes it hard to eradicate militant Islamism, as is no center of power you can knock out to end the war. However it also means Islamists are unable to concentrate the force required to really threaten us. At worst, they can kill a few thousand innocent civilians, which is of course horrible, but hardly on the same civilization-threatening level as Nazis exterminating millions or Communists threatening to eradicate Europe with nukes.
Second, unlike Communism and Fascism, militant Islamism has little attraction as an ideology in the West. During the cold war communist sympathizers infiltrated governments and other key institutions in the United States and Western Europe. Communist American scientists stole U.S military technology and helped Stalin build nuclear weapons. The same will not happen with regards to radical Islam. Today there are very few American scientists who are True Believers (Useful Idiots) and likely to steal nuclear secrets and give them to Bin Laden.
However what does annoy me is when the elite uses statistics to manipulate the public.
In Sweden the public is currently worried about Islamic terror, after two recent incidents with Islamist terrorist. Once again the elites, (media, politician and academics) have ganged up against the public and are trying to downplay the terrorist threat from militant Islam.
You see, while ordinary citizens may have gotten the impression from the nightly news that adherents of militant Islamism are statistically overrepresented in terms of international terrorism, the Enlightened Classes know better. The real reason the poor fools in the public believe there is a link between militant Islam and international terrorism is "islamophobia".
There was a debate in the parliament recently. Both the Social Democrats and the other parties in the left claimed that the main terrorist threat to the Swedish public today is not radical Islam, but right wing extremism...
The Green party representative explained that terrorism "is about emotional and social aspects rather than ideological ones".
This is the left, but the right is not much better.
So how in god's name would you convince the public of the preposterous claim that militant Islam is not the main terrorist threat in Europe? Their method is interesting, both in showing ingenuity in finding ways to trick the public, and stupidity in what they are willing to convince themselves of. The solution is namely to rely on the number of terrorist attacks, rather than on casualties from terrorism!
In Spain and Northern Ireland in particular, there are lots of tiny terrorist attacks by domestic terrorists every year. These attacks typically don't kill anyone, and often don't appear aimed at killing anyone. Characteristic examples from the RAND database:
"The headquarters of the Spanish Socialist Workers' Party (PSOE) in Baranain had several windows broken when it was attacked with stones. The damage is estimated at 50,000 pesetas."
"A group of radicals attacked a Renault dealership by setting a fire which damaged three vans belonging to the firm. A container of flammable liquids and firework rockets were used to start the fire."
Al-Qaida on the other hand specialized on a few spectacular attacks aimed at killing as many as possible.
Thus if we ignore the deaths and treat each attack as equal, you can show that Islamists commit a smaller number of terrorist acts than domestic terrorists. This method absurdly assumes that bringing down the Twin Towers on September 11 is equal to ETA vandalizing some property in Spain (each is one attack after all).
This is for instance what Sweden's largest daily DN does, using a report from Interpol about the number of terrorist incidences in Europe, and concluding that "Islamists terror attacks are unusual in Europe".
My calculations show that DN is misleading it's readers. As usual, when it comes to issues shrouded in political correctness, the public is better off relying on their own impression than "scientists" and "experts" in the media.
Thursday, February 17, 2011
The crisis of Social Democracy in Scandinavia
The Swedish Social Democrats are currently in disarray. A party that alone held power virtually uninterrupted for most of the 20th century currently only has support of about 25-28% of voters.
I am not going to offer them advice how to recover, since I am their ideological opponent, and I myself would never trust advice from my ideological opponents to be in my best interest.
Most of the explanations offered for the meltdown relies on uniquely Swedish events and recent phenomenon. However the exact same thing happened in Denmark and Norway. Moreover, the Swedish Social Democrats have been slowly declining for decades, which was obscured by a couple of odd victories following the 1990s financial crisis. So most likely the explanation for the decline has a systematic component and is common to all three Scandinavian countries.
One explanation is reduced class-consciousness among the working class, who are also becoming a smaller share of the population. There is not much the Labour parties can do about this.
Another explanation is resistance of working class voters to immigration. The Scandinavian working class have been hit hardest by the adverse impact of non-western immigration.
* First, working class neighborhoods have turned into ghettos, forcing them to either move out or live with high crime, troubled schools and other negative social externalities.
* Second, unskilled immigrants put downward pressure on wages and employment.
* Third, immigration costs billions of dollars, which means less money over to welfare state services.
* Lastly, the immigrants don't have Lutheran work ethics and strong social pressure not to abuse the welfare state like the Scandinavians. Many non-western immigrants take full advantage of all the generous benefits, and some cheat if they can. This behavior has forced the Scandinavians to make social insurance payments less generous for everybody, and to introduce harsher controls. The unintended consequence is that a 55 year old Swedish working class women with health problems cannot get early retirement as easily as she could in 1985, because the system has become less trusting to everyone due to abuse.
Working class voters are also less likely to benefit from immigration in the form of cheaper services (working class Scandinavians cannot afford maids), and unlike the middle class they don't even pretend to enjoy Iraqi and Albanian cultural expressions.
Instead of reforming policies, the reaction of the Social Democrats to failed integration has been to ratcheted up pro-immigration propaganda. As a consequence, many of the working class feel abandoned by the Social Democrats, and are in turn abandoning them in favor of populist anti-immigration parties.
These graphs illustrate the development since the start of the Era of Social Democratic Dominance in 1936.
In Denmark the Social Democrats are expected to make a come-back in the next election. This lends support to my hypothesis, since the Danish Social Democrats have gone the furthest in giving up on multiculturalism and mass-immigration.
I am not going to offer them advice how to recover, since I am their ideological opponent, and I myself would never trust advice from my ideological opponents to be in my best interest.
Most of the explanations offered for the meltdown relies on uniquely Swedish events and recent phenomenon. However the exact same thing happened in Denmark and Norway. Moreover, the Swedish Social Democrats have been slowly declining for decades, which was obscured by a couple of odd victories following the 1990s financial crisis. So most likely the explanation for the decline has a systematic component and is common to all three Scandinavian countries.
One explanation is reduced class-consciousness among the working class, who are also becoming a smaller share of the population. There is not much the Labour parties can do about this.
Another explanation is resistance of working class voters to immigration. The Scandinavian working class have been hit hardest by the adverse impact of non-western immigration.
* First, working class neighborhoods have turned into ghettos, forcing them to either move out or live with high crime, troubled schools and other negative social externalities.
* Second, unskilled immigrants put downward pressure on wages and employment.
* Third, immigration costs billions of dollars, which means less money over to welfare state services.
* Lastly, the immigrants don't have Lutheran work ethics and strong social pressure not to abuse the welfare state like the Scandinavians. Many non-western immigrants take full advantage of all the generous benefits, and some cheat if they can. This behavior has forced the Scandinavians to make social insurance payments less generous for everybody, and to introduce harsher controls. The unintended consequence is that a 55 year old Swedish working class women with health problems cannot get early retirement as easily as she could in 1985, because the system has become less trusting to everyone due to abuse.
Working class voters are also less likely to benefit from immigration in the form of cheaper services (working class Scandinavians cannot afford maids), and unlike the middle class they don't even pretend to enjoy Iraqi and Albanian cultural expressions.
Instead of reforming policies, the reaction of the Social Democrats to failed integration has been to ratcheted up pro-immigration propaganda. As a consequence, many of the working class feel abandoned by the Social Democrats, and are in turn abandoning them in favor of populist anti-immigration parties.
These graphs illustrate the development since the start of the Era of Social Democratic Dominance in 1936.
In Denmark the Social Democrats are expected to make a come-back in the next election. This lends support to my hypothesis, since the Danish Social Democrats have gone the furthest in giving up on multiculturalism and mass-immigration.
Sunday, February 13, 2011
Are we worse off than in 1973?
Tyler Cowen has a new book out, called The Great Stagnation. It is reviewed by anti-immigration pundit Steve Sailer.
Sailer makes a familiar argument, which is that while the rich are getting richer, ordinary Americans have experienced no improvement in their standard of living since 1973, and perhaps even a detritions (1973 is used because that was the height of the post-war boom).
He argues that while entertainment and plastic toys have become cheaper, the things that provide deep life satisfaction for the middle and working classes have become more expensive. This includes a nice home in a safe neighborhood and college for your children.
It is common for free-market economists to dismiss this line of reasoning. I don't want to do that, so let me note from the beginning that:
* While technological change, unskilled immigration and trade/outsourcing have helped the upper half, they have likely hurt working class Americans.
* The standard of living was improving rapidly for all between 1946-1973, whereas 1973-2008 at best provided slow and uneven improvement.
* When measuring standard of living, it's not enough to look at the price of consumer products such as laptops. The cost of keeping yourself and your children socially in the middle class - for example sending them to college or even having a stay at home mother - should be included.
However, I disagree with the claim that things are worse than 1973. This is based on relying on one data-set, which is the median hourly wage for non-supervisory production workers as measured by BLS, inflation adjusted by the CPI. But these wages for in total account for less than 40% of national income, and CPI-adjustment is not perfect.
Other data-series tell us a less pessimistic story.
Adjusting for inflation, the Census Bureau measure of median household income increased by 10% between 1973-2008. This is a broader and in my view better measure of income than BLS wages.
Contrary to popular perception, aggregate hours worked per adult are no higher than in 1973. Furthermore, this comparison does not take into account changing demographics. Compared to 1973, America has taken in millions of unskilled Hispanic workers, who earn less and depress the median. If we look at non-Hispanic white households, real median income increased by 15%. For African Americans, real median household income increased by 22%.
Sailer, liberals and paleo-conservatives are convinced that the CPI under-estimates inflation (so that using CPI over-estimates growth). However the Boskin Commission which studied the CPI carefully concluded that the CPI massively over-estimates inflation (and thus leads to underestimation of growth). This is because CPI cannot fully measure technology driven quality improvement, the value of completely new products, and cheaper outlets such as Wal-Mart.
Here is one area where I am not willing to back off even one inch from economist-conventional wisdom. On inflation of consumer goods, Boskin is right and Sailer is wrong.
Another data point is the Survey of Consumer Finance, which measures wealth. Real Median household wealth was $40.000 in 1970 and $88.000 in 2009 (after the crash).
You might argue that it is becoming cheaper to buy "stuff", but more expensive to buy truly "important things" like housing, health care, education for your children, and that the latter matters more for your well-being than I-pods. But people still spend a huge share of their income on "stuff", before they spend on "important things". Here I rely on The Consumer Expenditure Survey. Unfortunately it doesn't go back to 1973, so I will use the last available comparable year, 1984.
I will look at the middle 20% of the population. One result that jumps at you is that the middle class now pays much lower taxes, taxes are down by about $2.000 per household.
Let's define "stuff" as food, clothing and services, transportation, utilities, fuels, and public services, household furnishings and equipment, housekeeping supplies, personal care products and services and entertainment.
The price these has decreased, and the quality improved. In 1984 "stuff" was 62% of expenditure, compared to 52% in 2009. In absolute numbers, it declined from $25.000 to $21.000. Of course $21.000 in real dollars today for instance buys you a better car than in 1973.
Expenditure on "important things", which is housing, health care and education, increased from 37% to 45%. In absolute numbers it went from $15.000 to $18.000. This underestimates the increase, because employer provided health care is also paid through forgone wages. Still, since people spend such a large share of income on it, you can't just dismiss "stuff" when discussing the quality of life. Even if you believe that it is unimportant, reduced cost of "stuff" means people can more money over to spend on the things they truly care about.
I have also calculated my preferred comparison of income between 1973 and 2008. This relies on measures of aggregate personal income from National Account, calculated by the Bureau of Economic Analysis. These figures have a broader definition of income (I believe the broadest available), and a different inflation adjustment.
I have used measures of the share of national income that goes to the rich from Saez and Pickety. The ten percent richest were those who, in real dollars, earned more than $86,000 in 1973 and more than $112,000 in 2008. I remove the share of income that goes to the ten percent who make more than this. I call the remaining 90% of the population "Lower and Middle classes".
Per capita national income closely follows the rate of GDP growth. Between 1946-1973, overall per capita income increased by 2.4% annually, and 2.6% for the Lower and Middle classes.
Between 1973-2008 per capita income increased by 1.9% per year. This confirms that Tyler Cowen is right about lower growth rate than the Post-War Golden Age, but also that people who suggest zero growth are wrong. However inequality increased during these years, so the per capita income of the Lower and Middle classes only increased by 1.1% per year. Still, 1.1% is not zero, the per capita personal income of the Lower and Middle classes is now one and a half time higher compared to 1973, adjusted for inflation.
Lastly Sailer anecdotally discusses the cost of going to college and buying a home when he was young.
The Median home in 2010 dollars went from $160.000 in the 1970s to $221.000 in 2010. This does not guarantee that the quality is the same, but overall prices have only increased somewhat.
I looked up the cost of higher education from the College Board.
In 1980, Full Tuition plus Room and Board cost $14,700 in Private College and $6,700 in public college (in 2010 dollars). In 2010, the figures were $37,000 and $16,100. For my non-American readers it could be important to keep in mind that most students don't pay full tuition, that this does not include all the public subsidies for higher education, and that two thirds of Americans go to public universities.
So if you have two kids, and send one to public and one to private school, and pay for everything in undergraduate (they will borrow for graduate school themselves), the real cost went from $86.000 in 1980 to $213.000 in 2010. If we add the increase of the cost of a median house, this is close to $200.000 more a middle class family has to pay to stay in the same place they were four decades ago. Or force their children to do what I did, borrow yourself for your own cost of education.
I find a mixed picture, but also pretty strong arguments that "ordinary people are worse off than 1973" is not quite true. In particular, I want to warn people that the BLS median hourly wage figure is not the only measure of income, and likely not the best. It would be more fair to say that ordinary people are worse off compared to what they had a right to expect. They are perhaps also worse off compared to what alternative economic policies would have resulted in, most obvious being a high-wage policy which limits unskilled immigration.
Sailer makes a familiar argument, which is that while the rich are getting richer, ordinary Americans have experienced no improvement in their standard of living since 1973, and perhaps even a detritions (1973 is used because that was the height of the post-war boom).
He argues that while entertainment and plastic toys have become cheaper, the things that provide deep life satisfaction for the middle and working classes have become more expensive. This includes a nice home in a safe neighborhood and college for your children.
It is common for free-market economists to dismiss this line of reasoning. I don't want to do that, so let me note from the beginning that:
* While technological change, unskilled immigration and trade/outsourcing have helped the upper half, they have likely hurt working class Americans.
* The standard of living was improving rapidly for all between 1946-1973, whereas 1973-2008 at best provided slow and uneven improvement.
* When measuring standard of living, it's not enough to look at the price of consumer products such as laptops. The cost of keeping yourself and your children socially in the middle class - for example sending them to college or even having a stay at home mother - should be included.
However, I disagree with the claim that things are worse than 1973. This is based on relying on one data-set, which is the median hourly wage for non-supervisory production workers as measured by BLS, inflation adjusted by the CPI. But these wages for in total account for less than 40% of national income, and CPI-adjustment is not perfect.
Other data-series tell us a less pessimistic story.
Adjusting for inflation, the Census Bureau measure of median household income increased by 10% between 1973-2008. This is a broader and in my view better measure of income than BLS wages.
Contrary to popular perception, aggregate hours worked per adult are no higher than in 1973. Furthermore, this comparison does not take into account changing demographics. Compared to 1973, America has taken in millions of unskilled Hispanic workers, who earn less and depress the median. If we look at non-Hispanic white households, real median income increased by 15%. For African Americans, real median household income increased by 22%.
Sailer, liberals and paleo-conservatives are convinced that the CPI under-estimates inflation (so that using CPI over-estimates growth). However the Boskin Commission which studied the CPI carefully concluded that the CPI massively over-estimates inflation (and thus leads to underestimation of growth). This is because CPI cannot fully measure technology driven quality improvement, the value of completely new products, and cheaper outlets such as Wal-Mart.
Here is one area where I am not willing to back off even one inch from economist-conventional wisdom. On inflation of consumer goods, Boskin is right and Sailer is wrong.
Another data point is the Survey of Consumer Finance, which measures wealth. Real Median household wealth was $40.000 in 1970 and $88.000 in 2009 (after the crash).
You might argue that it is becoming cheaper to buy "stuff", but more expensive to buy truly "important things" like housing, health care, education for your children, and that the latter matters more for your well-being than I-pods. But people still spend a huge share of their income on "stuff", before they spend on "important things". Here I rely on The Consumer Expenditure Survey. Unfortunately it doesn't go back to 1973, so I will use the last available comparable year, 1984.
I will look at the middle 20% of the population. One result that jumps at you is that the middle class now pays much lower taxes, taxes are down by about $2.000 per household.
Let's define "stuff" as food, clothing and services, transportation, utilities, fuels, and public services, household furnishings and equipment, housekeeping supplies, personal care products and services and entertainment.
The price these has decreased, and the quality improved. In 1984 "stuff" was 62% of expenditure, compared to 52% in 2009. In absolute numbers, it declined from $25.000 to $21.000. Of course $21.000 in real dollars today for instance buys you a better car than in 1973.
Expenditure on "important things", which is housing, health care and education, increased from 37% to 45%. In absolute numbers it went from $15.000 to $18.000. This underestimates the increase, because employer provided health care is also paid through forgone wages. Still, since people spend such a large share of income on it, you can't just dismiss "stuff" when discussing the quality of life. Even if you believe that it is unimportant, reduced cost of "stuff" means people can more money over to spend on the things they truly care about.
I have also calculated my preferred comparison of income between 1973 and 2008. This relies on measures of aggregate personal income from National Account, calculated by the Bureau of Economic Analysis. These figures have a broader definition of income (I believe the broadest available), and a different inflation adjustment.
I have used measures of the share of national income that goes to the rich from Saez and Pickety. The ten percent richest were those who, in real dollars, earned more than $86,000 in 1973 and more than $112,000 in 2008. I remove the share of income that goes to the ten percent who make more than this. I call the remaining 90% of the population "Lower and Middle classes".
Per capita national income closely follows the rate of GDP growth. Between 1946-1973, overall per capita income increased by 2.4% annually, and 2.6% for the Lower and Middle classes.
Between 1973-2008 per capita income increased by 1.9% per year. This confirms that Tyler Cowen is right about lower growth rate than the Post-War Golden Age, but also that people who suggest zero growth are wrong. However inequality increased during these years, so the per capita income of the Lower and Middle classes only increased by 1.1% per year. Still, 1.1% is not zero, the per capita personal income of the Lower and Middle classes is now one and a half time higher compared to 1973, adjusted for inflation.
Lastly Sailer anecdotally discusses the cost of going to college and buying a home when he was young.
The Median home in 2010 dollars went from $160.000 in the 1970s to $221.000 in 2010. This does not guarantee that the quality is the same, but overall prices have only increased somewhat.
I looked up the cost of higher education from the College Board.
In 1980, Full Tuition plus Room and Board cost $14,700 in Private College and $6,700 in public college (in 2010 dollars). In 2010, the figures were $37,000 and $16,100. For my non-American readers it could be important to keep in mind that most students don't pay full tuition, that this does not include all the public subsidies for higher education, and that two thirds of Americans go to public universities.
So if you have two kids, and send one to public and one to private school, and pay for everything in undergraduate (they will borrow for graduate school themselves), the real cost went from $86.000 in 1980 to $213.000 in 2010. If we add the increase of the cost of a median house, this is close to $200.000 more a middle class family has to pay to stay in the same place they were four decades ago. Or force their children to do what I did, borrow yourself for your own cost of education.
I find a mixed picture, but also pretty strong arguments that "ordinary people are worse off than 1973" is not quite true. In particular, I want to warn people that the BLS median hourly wage figure is not the only measure of income, and likely not the best. It would be more fair to say that ordinary people are worse off compared to what they had a right to expect. They are perhaps also worse off compared to what alternative economic policies would have resulted in, most obvious being a high-wage policy which limits unskilled immigration.
Tuesday, February 8, 2011
America wrong continent for High-Speed Trains
Today the White House released a plan to invest anther $53 billion in High-Speed rail.
The New York Times headlines this "U.S. Plays Catch-Up on High-Speed Rail", admiring High-Speed trains in China and Europe. Basically, the American Left argues that since Western Europe and China have high-speed rail, and since they believe that Western Europe and China have better economic policy than the United States, we should emulate them and build fast trains.
I often argue that European style policies will not work in America because of demographics and cultural differences. I can understand that not all readers are convinced that Americans are that different from Europeans. However, I hope every reader accepts that the U.S is geographically different from Europe and Asia.
High-Speed train countries Spain and France have 3 times higher population density than America. China has 4 times higher, Germany 7 times higher, Japan 10 times higher, South Korea 15 times higher and Taiwan 20 times higher population density than the U.S. Germany is more densely populated than New York state, and China more densely populated than California.
Countries that like America have a lot land compared to people, such as Canada, Scandinavia, Russia and Australia have not made any large scale investments in high-speed trains.
Let me illustrate this graphically. I take the total high-speed miles from The International Union of Railways, and plot the density of the high-speed-rail network with population density.
The United States is not an outlier as the White-House suggests, the U.S is exactly where our population density would predict. Only after President Obama's plan will the U.S become a outlier, a country with more High-Speed Train that population density would predict (the figure after Obama's plan is my estimate based on White House material).
High-Speed trains are not only expensive, they are slow when compared to air-travel. Take one of the least crazy high-speed train projects, connecting Los Angeles and San Francisco. The White House estimates are that this trip will take 2 hours 40 minutes. The same trip by commercial flight takes 1 hours 20 minutes. Even if you add an extra one hour for security check, the trip is faster by air (you also have to drive to the airport, but the same is true for trains).
After the first terrorist attack against high-speed trains, the security advantage would diminish. If we really wanted to and had an extra $53 billion over, we could invest in flying faster, in making the security process more effective, or (most sensibly) improving the high-way system.
Another fact Liberals ignore is that air-travel is cheaper in the U.S, costing about half per mile of what it does in Europe (perhaps due to economies of scale and higher competitiveness).
Investing in High-Speeds trains is likely a "White Elephant", a massive visible project that gets politicians attention, but is a bad deal for tax-payers. I hope we are not building it just to fulfill juvenile fantasies of making the U.S more like Europe.
The New York Times headlines this "U.S. Plays Catch-Up on High-Speed Rail", admiring High-Speed trains in China and Europe. Basically, the American Left argues that since Western Europe and China have high-speed rail, and since they believe that Western Europe and China have better economic policy than the United States, we should emulate them and build fast trains.
I often argue that European style policies will not work in America because of demographics and cultural differences. I can understand that not all readers are convinced that Americans are that different from Europeans. However, I hope every reader accepts that the U.S is geographically different from Europe and Asia.
High-Speed train countries Spain and France have 3 times higher population density than America. China has 4 times higher, Germany 7 times higher, Japan 10 times higher, South Korea 15 times higher and Taiwan 20 times higher population density than the U.S. Germany is more densely populated than New York state, and China more densely populated than California.
Countries that like America have a lot land compared to people, such as Canada, Scandinavia, Russia and Australia have not made any large scale investments in high-speed trains.
Let me illustrate this graphically. I take the total high-speed miles from The International Union of Railways, and plot the density of the high-speed-rail network with population density.
The United States is not an outlier as the White-House suggests, the U.S is exactly where our population density would predict. Only after President Obama's plan will the U.S become a outlier, a country with more High-Speed Train that population density would predict (the figure after Obama's plan is my estimate based on White House material).
High-Speed trains are not only expensive, they are slow when compared to air-travel. Take one of the least crazy high-speed train projects, connecting Los Angeles and San Francisco. The White House estimates are that this trip will take 2 hours 40 minutes. The same trip by commercial flight takes 1 hours 20 minutes. Even if you add an extra one hour for security check, the trip is faster by air (you also have to drive to the airport, but the same is true for trains).
After the first terrorist attack against high-speed trains, the security advantage would diminish. If we really wanted to and had an extra $53 billion over, we could invest in flying faster, in making the security process more effective, or (most sensibly) improving the high-way system.
Another fact Liberals ignore is that air-travel is cheaper in the U.S, costing about half per mile of what it does in Europe (perhaps due to economies of scale and higher competitiveness).
Investing in High-Speeds trains is likely a "White Elephant", a massive visible project that gets politicians attention, but is a bad deal for tax-payers. I hope we are not building it just to fulfill juvenile fantasies of making the U.S more like Europe.
Sunday, February 6, 2011
Blue State and Red State IQ is identical
The American left often congratulates itself on being smarter than the right. On the surface, this claim sounds plausible, since academics are disproportionally Democrat, and since some of the states most associated with poor educational outcomes (such as Mississippi) are solidly Republican.
The first problem with this line of argumentation is that it is lazy. The assumption is that if smart people vote Democrat, that in out of itself proves Democrats are right, without needing to prove the Republicans wrong in substance. However, like everyone else, the educated vote based on class interests, cultural identification and the ideological atmosphere of college, not just cold rational analysis of policy differences.
Second, there is actually no scientific evidence for the claim that Democrats are smarter, since no American political science database of voting to my knowledge collects IQ-data. Data on education and voting shows that Democrats and Republicans have identical average years of education. Democrats are over-represented among the most educated but also among the least educated, while Republicans have the middle.
Third, these voting patterns are constantly shifting, and differ across countries (and even within U.S states). As late as 1988 the educated voted overwhelmingly Republican. Through friends in Sweden have access to a high-quality dataset of IQ and voting, which confirms that Swedish Social Democrat voters have lower test scores than Swedish conservative voters. The fundamental arguments between pro-market and pro-government are the same in Europe and the United States. But the American left would never accept that the European right has superior arguments just because their voters have higher IQ than the European left (nor should anyone accept this premise). Yet the same liberals try to bully people in the U.S to vote for them based on the claim that Democrats are smarter than Republicans.
To make things worse, unlike Europe the American left have not taken the time to demonstrate their claim using actual test-score data. Democrats instead simple assume that they are smarter than everyone else!
This theory can ultimately only be evaluated using individual level data. The next best thing is analyzing proxies for state I.Q. Let me caution that this is problematic. Regarding income Andrew Gelman and others have shown the paradox that while Democrat States are on average richer than Republican States, Democrat voters are on average poorer than Republican voters.
Part of the explanation is that in Red states the poor vote Democrat and the rich Republican. Uneducated whites in the South, the demographic group the media likes to paint as bigoted Republicans, actually went close to 45% for Obama, perhaps because they earn little and didn't mind a little redistribution. McCain's margin in the South came from educated whites, a detail the media prefers to ignore since it contradicts their stereotypes.
I will use this paper in the prestigious journal Intelligence to get estimates of state I.Q, based on NAEP test scores.
If you just glance at the list, you might get the impression that the Blue states are indeed smarter than the Red states. I define Red states based on the 2004 Bush map, so New Hampshire is defined as blue and New Mexico as Red.
4 out of the top 5 are Democrat:
1. Massachusetts (D)
2. New Hampshire (D)
3. North Dakota (R)
4. Vermont (D)
5. Minnesota (D)
And 3 out of the 5 bottom states are Republican. Although the author doesn't include the District of Columbia, which would have had the bottom slot if included.
46. New Mexico (R)
47. Hawaii (D)
48. California (D)
49. Louisiana (R)
50. Mississippi (R)
However, looking more carefully at the list, the results are not that consistent. Red Midwestern and Mountain states such as the Dakotas, Montana, Kansas, Nebraska, Wyoming and Idaho do quite well. Meanwhile, large blue states such as California, Maryland and Illinois have unimpressive scores. So let's just estimate the average for all the Red states, and all the Blue states, weighted by the population of each state.
The result is not really surprising: The average I.Q level of Blue and Red States is identical: Blue States have an average IQ of 99.5 and Red States have an average I.Q of 99.4.
It would be strange if we found large differences, since we know from exit polls that Republican voters are on average no less educated than Democrat voters, and since I.Q and education are strongly correlated.
If the District of Columbia is included, the Red States would score slightly higher than the Blue states (although with I.Q a 1 point difference is meaningless).
Unless Democrats can produce a micro-study of adult voters that demonstrates that Democrat voters have significantly higher I.Q than Republican voters, they should respect the state of science and stop spreading the myth that they are smarter than everyone else.
P.S.
Here is the full list of states and estimated IQ:
1. Massachusetts 104.3
2. New Hampshire 104.2
3. North Dakota 103.8
4. Vermont 103.8
5. Minnesota 103.7
6. Maine 103.4
7. Montana 103.4
8. Iowa 103.2
9. Connecticut 103.1
10. Wisconsin 102.9
11. Kansas 102.8
12. New Jersey 102.8
13. South Dakota 102.8
14. Wyoming 102.4
15. Nebraska 102.3
16. Virginia 101.9
17. Washington 101.9
18. Ohio 101.8
19. Indiana 101.7
20. Colorado 101.6
21. Pennsylva. 101.5
22. Idaho 101.4
23. Oregon 101.2
24. Utah 101.1
25. Missouri 101
26. New York 100.7
27. Michigan 100.5
28. Delaware 100.4
29. N. Carolina 100.2
30. Texas 100
31. Illinois 99.9
32. Maryland 99.7
33. Rhode Island99.5
34. Kentucky 99.4
35. Oklahoma 99.3
36. Alaska 99
37. W. Virginia 98.7
38. Florida 98.4
39. S. Carolina 98.4
40. Georgia 98
41. Tennessee 97.7
42. Arkansas 97.5
43. Arizona 97.4
44. Nevada 96.5
45. Alabama 95.7
46. New Mexico 95.7
47. Hawaii 95.6
48. California 95.5
49. Louisiana 95.3
50. Mississippi 94.2
The first problem with this line of argumentation is that it is lazy. The assumption is that if smart people vote Democrat, that in out of itself proves Democrats are right, without needing to prove the Republicans wrong in substance. However, like everyone else, the educated vote based on class interests, cultural identification and the ideological atmosphere of college, not just cold rational analysis of policy differences.
Second, there is actually no scientific evidence for the claim that Democrats are smarter, since no American political science database of voting to my knowledge collects IQ-data. Data on education and voting shows that Democrats and Republicans have identical average years of education. Democrats are over-represented among the most educated but also among the least educated, while Republicans have the middle.
Third, these voting patterns are constantly shifting, and differ across countries (and even within U.S states). As late as 1988 the educated voted overwhelmingly Republican. Through friends in Sweden have access to a high-quality dataset of IQ and voting, which confirms that Swedish Social Democrat voters have lower test scores than Swedish conservative voters. The fundamental arguments between pro-market and pro-government are the same in Europe and the United States. But the American left would never accept that the European right has superior arguments just because their voters have higher IQ than the European left (nor should anyone accept this premise). Yet the same liberals try to bully people in the U.S to vote for them based on the claim that Democrats are smarter than Republicans.
To make things worse, unlike Europe the American left have not taken the time to demonstrate their claim using actual test-score data. Democrats instead simple assume that they are smarter than everyone else!
This theory can ultimately only be evaluated using individual level data. The next best thing is analyzing proxies for state I.Q. Let me caution that this is problematic. Regarding income Andrew Gelman and others have shown the paradox that while Democrat States are on average richer than Republican States, Democrat voters are on average poorer than Republican voters.
Part of the explanation is that in Red states the poor vote Democrat and the rich Republican. Uneducated whites in the South, the demographic group the media likes to paint as bigoted Republicans, actually went close to 45% for Obama, perhaps because they earn little and didn't mind a little redistribution. McCain's margin in the South came from educated whites, a detail the media prefers to ignore since it contradicts their stereotypes.
I will use this paper in the prestigious journal Intelligence to get estimates of state I.Q, based on NAEP test scores.
If you just glance at the list, you might get the impression that the Blue states are indeed smarter than the Red states. I define Red states based on the 2004 Bush map, so New Hampshire is defined as blue and New Mexico as Red.
4 out of the top 5 are Democrat:
1. Massachusetts (D)
2. New Hampshire (D)
3. North Dakota (R)
4. Vermont (D)
5. Minnesota (D)
And 3 out of the 5 bottom states are Republican. Although the author doesn't include the District of Columbia, which would have had the bottom slot if included.
46. New Mexico (R)
47. Hawaii (D)
48. California (D)
49. Louisiana (R)
50. Mississippi (R)
However, looking more carefully at the list, the results are not that consistent. Red Midwestern and Mountain states such as the Dakotas, Montana, Kansas, Nebraska, Wyoming and Idaho do quite well. Meanwhile, large blue states such as California, Maryland and Illinois have unimpressive scores. So let's just estimate the average for all the Red states, and all the Blue states, weighted by the population of each state.
The result is not really surprising: The average I.Q level of Blue and Red States is identical: Blue States have an average IQ of 99.5 and Red States have an average I.Q of 99.4.
It would be strange if we found large differences, since we know from exit polls that Republican voters are on average no less educated than Democrat voters, and since I.Q and education are strongly correlated.
If the District of Columbia is included, the Red States would score slightly higher than the Blue states (although with I.Q a 1 point difference is meaningless).
Unless Democrats can produce a micro-study of adult voters that demonstrates that Democrat voters have significantly higher I.Q than Republican voters, they should respect the state of science and stop spreading the myth that they are smarter than everyone else.
P.S.
Here is the full list of states and estimated IQ:
1. Massachusetts 104.3
2. New Hampshire 104.2
3. North Dakota 103.8
4. Vermont 103.8
5. Minnesota 103.7
6. Maine 103.4
7. Montana 103.4
8. Iowa 103.2
9. Connecticut 103.1
10. Wisconsin 102.9
11. Kansas 102.8
12. New Jersey 102.8
13. South Dakota 102.8
14. Wyoming 102.4
15. Nebraska 102.3
16. Virginia 101.9
17. Washington 101.9
18. Ohio 101.8
19. Indiana 101.7
20. Colorado 101.6
21. Pennsylva. 101.5
22. Idaho 101.4
23. Oregon 101.2
24. Utah 101.1
25. Missouri 101
26. New York 100.7
27. Michigan 100.5
28. Delaware 100.4
29. N. Carolina 100.2
30. Texas 100
31. Illinois 99.9
32. Maryland 99.7
33. Rhode Island99.5
34. Kentucky 99.4
35. Oklahoma 99.3
36. Alaska 99
37. W. Virginia 98.7
38. Florida 98.4
39. S. Carolina 98.4
40. Georgia 98
41. Tennessee 97.7
42. Arkansas 97.5
43. Arizona 97.4
44. Nevada 96.5
45. Alabama 95.7
46. New Mexico 95.7
47. Hawaii 95.6
48. California 95.5
49. Louisiana 95.3
50. Mississippi 94.2
Subscribe to:
Posts (Atom)