Sunday, January 31, 2010

The Bush Tax Cuts Worked.

The ”Bush tax cuts on the rich” are generally viewed as a “failure”. If the criterion is that the tax cuts would have saved the economy from financial meltdown, that is certainly true. Of course it is absurd to think that cutting taxes by 3-5 percentage points would prevent a gigantic financial meltdown.

Another question is if the tax cuts were entirely self-financing, so that cutting taxes would completely pay for itself. This did not happen, nor should we have expected it to given what we know about short-medium run responsiveness of the tax base to tax rates.

However, this requirement is also too extreme. I think the important question is if tax cuts stimulate growth by a reasonable amount. For tax cuts to be completely self-financing, they have to stimulate growth enourmosly. This only happens either if the responsiveness to taxes is very high (in fact it seems to be moderately high) or if tax rates are extremly high (which they are not in the US.). But what if tax cuts in the American setting stimulates growth by a good deal, but not enough to be self-financing? Should we throw away a useful tool just because it is not Voodoo?

The detractors of the tax cuts seem to be going to the other extreme, arguing that the Bush tax cuts had no effect at all on growth, and that they were not at all self-financing, and therefore that supply-side economics is totally wrong.

But remember that actual supply side economics claims that tax cuts stimulate growth. It is only vulgar or straw-man supply side economics that claim that tax cuts always stimulate growth by the extreme amounts required for tax cuts to be 100% (or more) self financing.

Here research comes in. Economists have in fact studied the effects of the “Bush tax cuts for the rich” on the tax base. The answer is that they did stimulate the economy, and were partially self-financing, about 40% self-financing to be exact. That is a pretty good deal: for each $0.6 dollars that the government loses in revenue the private sector gains $1 dollars.

The paper “The 2001 and 2003 Tax Rate Reductions: An Overview and Estimate of the Taxable Income Response” By Gerald Auten, Robert Carroll and Geoffrey Gee in the National Tax journal in 2008 calculates the responsiveness of income on tax rates, and finds that also in this case did people whose tax rates go down increase their taxable income (their standard estimate of the elasticity of taxable income if 0.4 in this period, in line with the literature). Regarding the tax cuts for the rich, they find that:

“Overall, the increase in taxable income translates into higher revenues that offset about 39 percent of the static revenue loss associated with the reduction in the top two tax rates.”


Empirical evidence suggests that Supply side economics worked as predicted in theory, also regarding the Bush tax cuts.

Friday, January 29, 2010

Revealed Preferences on Capitalism vs. the Welfare State


The American left portrays the Nordic countries as welfare-state Utopias. Swedes may earn less than Americans, but because of extensive public programs the quality of life is much higher.

We can have endless subjective debate about which system is better, and where the quality of life is higher (crime versus freedom, money versus equality etc.). Not to get stock in subjective measures, what does objective behavior show?

It would be no challenge if we were to compare the U.S with the average welfare state, such as France, U.K or Italy. The Nordic states are however the most successful welfare states, the examples the American left likes to put forward. (incidentally this would be like me only using Colorado, Minnesota and Connecticut when comparing social and economic outcomes of the U.S versus Europe, but let us accept the challenge).

In my judgment it is roughly equally easy to move from the U.S to Scandinavia that it is to move from Scandinavia to America. If you really want to you can do it. You just need to find a job or find a partner (and it is quite easy to move a few years in order to study and somehow stay). Arguably it is easier to move from the U.S to Scandinavia, since there are fewer restrictions on work visas, and since higher education is free (I should say tax financed). On the other hand it is harder to Americans to learn the language (even though Americans don't actually need to speak Swedish in Sweden, since everyone speaks some English). Let us be generous and say that it is equally easy either way.

Keep in mind that these figures are recent immigration, people who are actually born in the U.S or Sweden, Denmark, Norway, Finland or Iceland that have chosen to migrate, not about ancestry.

According to the latest figures, there were 36.000 Americans in the five Nordic countries. According to the 2000 census, there are 140.000 people born in the five Nordic countries that have migrated to America.

So almost 4 times as many people have moved from the welfare state to capitalism than from capitalism to the welfare state. Voting with their feet, people choice America over the Nordic welfare states.



How would the left respond to this fact? I guess they could say something along the line that the American system is worse for the low-skilled, and the Nordic system worse for the high-skilled. Low-skilled Americans are however less likely to be entrepreneurial enough to move to another country to improve their lives, whereas high skilled Scandinavians seek out better opportunity in America.

However remember that the left typically argues that the welfare state is better for all society, including the high skilled! Furthermore, 60% of the people who have moved from the Nordic countries to America have no college degrees.

In economics revealed preferences are considered more reliable than cheap talk, and revealed preferences in this case support capitalism over the welfare state.

Wednesday, January 27, 2010

Some spurious regressions

This graph is one reason I don't believe in cross-country comparisons of self-reported Happiness and Life Satisfaction. It compares an international index of happiness and life satisfaction with the gini-coefficient of inequality. As you can see, the relationship between happiness and inequality is positive (barely statically significant, and clearly of the wrong sign).

I think inequality is negatively related to true well being. This relation is partially direct, and more importantly emerges as almost all social problems, be they drug use, crime, high school dropout rate, or teen pregnancy, create income inequality. As a side note this second effect leads to an endless spurious correlations as the left claims that it is income differences that is causing crime or drug use, and demand income transfers as useless remedy for deeper social pathologies.

Self-reported happiness is mainly a ratio of expectations and outcome, not an absolute measure of well-being. That is probably why the relative self-reported happiness of women has fallen, while their status has increased. They are better off, but their expectations on life have risen even faster than their status in society, so their self-reported happiness has declined.

Furthermore, reported happiness seems to depend on culture. Latin Americans report high levels of happiness, whereas for example the French give low scores (perhaps the French believe that claiming to be happy about life is low-brow). My best guess is that this odd positive relationship between inequality and happiness is created by chance, as cultural areas with high inequality (Latin America, Arab countries) tend to give high answers on self-reported happiness.

I simply cannot accept that El Salvador is better off than in France, South Africa better than off than Hong Kong, or Guatemala better off than Germany. Revealed preferences show that per capita GDP dominates indices of self-reported happiness. When people get to choose they try to move from El Salvador to the US and from Nigeria to France, as GDP predicts, but opposed to what self-reported happiness would predict.

Those on the left, who (having lost the income creating battle) now want to replace objective measures such as GDP with subjective measures such as self-reported happiness should keep these problems in mind.

Saturday, January 23, 2010

Immigration from Haiti is a terrible way to help

A former Bush official argues in the Washington Post that the US should increase immigration from Haiti, in order to help Haiti as immigrants send home remittances. I will try to demonstrate that this policy is extremely inefficient from the point of view of the U.S., and may be directly harmful from the point of view of Haiti.

1. According to the
World Bank Haiti in 2008 got $1.3 billion in remittances. Let's assume generously that 80% of remittances come from the U.S (the world bank's own estimate is slightly less than this). Based on the latest American Community Survey there are 786,000 people of Haitian origins in the U.S. This means that each Haitian on average sends home $1,300 per year. This figure will surely go up temporarily due to the earthquake, but presumably go back down again, as happened after the 1994 crisis.

The most reliable estimate of the fiscal impacts of immigration was done by the prestigious
National Research Council, NAC (the research arm of the National Academy of Sciences, NAS).

Low skilled immigrants earn less than the average, pay less in taxes and receive more in public services such as health care, public housing, income aid etc. The NAC estimate is that the total net cost of each low-skilled immigrant for the US. State is
$120,000 in 2009 dollars. (High skilled immigrants in contrast are a net fiscal benefit for the U.S).

These figures may underestimate the costs. Since this study was made the costs of welfare services to lower income people has further expanded, especially Medicaid and S-CHIP, and may go further yet..


Haitian immigrants should be expected to have on average even lower education levels than the NAC study assumed for the low-skilled. Haiti is the country in the western hemisphere with the lowest education level. According to the
U.N 40% of the adult population is illiterate. Even if all Haiti's highly educated population moves, any large scale migration will contain overwhelming low-skilled immigrants. The per capita income of Haitians already in the U.S is only half the American average, Haitians are not (let's hope not yet) an economically successful group.

Generously using the figure for the merely low educated (whereas many Haitian have no education),
we are trading of a $120,000 cost for the U.S taxpayer per Haitian immigrant for yearly remittances of $1,300 dollars.

It would be cheaper for the American taxpayer to directly increase aid to Haiti, even put $120 billion in a bank account and give the interest to Haiti, rather than to take in another million Haitian immigrants and bear the inescapable fiscal burden of a low-skilled group.


Even if we absurdly assume each Haitian lives forever and sends remittances home forever, discounted at a 5% interest rate $1,300 per year is worth only
$26,000 compared to a cost for the U.S taxpayer of $120,000.

Immigration is an extremely blunt instrument to help Haiti, one which entails several dollars in fiscal
costs for the American taxpayer for each dollar that actually reaches Haiti. The rational policy from the point of view of an America trying to help Haiti is aid, not immigration.

2. The immigration policy is questionable also from Haiti's point of view. First of all direct aid is more valuable than the same amount spent on net benefits for immigrants to obtain remittances (remember also that at some point the immigrant group inevitably loses interest in Haiti and stops sending money). Even one
million immigrants would only be 5 year of current population growth for Haiti, and would make only a small dent in their overpopulation.

In other more subtle ways emigration could be disastrous for Haiti. Even though most of the emigrants are low-skilled, historical experience indicates that the Haitian elite is more likely to emigrate. The highly educated Haitian emigrants are simply too few to make a noticeable positive impact on the economy of the U.S, but such a huge share of the of Haiti's tiny high-skilled group that they make a sizable negative impact on Haiti.


There are currently 75,000 Haitian immigrants in the U.S. with a college degrees. I would guess the proportions in Canada are similar. According to this
study that analyzes the Haiti Living Conditions Survey only 1.36% of adults in Haiti have Tertiary education. This means that that there are already more Haitians with college degrees in the US and Canada than there are left in Haiti!

Do we really believe that this exodus of the educated, of whom Haiti had so few to begin with, has benefited Haiti?
Should we encourage this further?

The extreme poverty of Haiti has been discussed recently. I don't have any empirical evidence for this but economic theory predicts that having half the educated population leave for the West may be a part of the puzzle for why Haiti remains so poor.


Proposing generous large scale immigration may make commentators feel good about themselves but careful analysis shows that it makes no sense, either from Americans' or Haiti's
point of view.

Friday, January 22, 2010

A Democratic deficit in Sweden

I mentioned in the last post that the Swedish Social Democrats were in power 85% of the time between 1932-2010. Sweden is a Social Democratic country. What few people realize is however that the complete dominance in power for the left was achieved with just a bare majority of the votes.

Between 1932-2010, the Social Democrats received on average 44.3% of the vote. Looking at the two block-vote (and excluding "other" parties outside parliament) the Social Democrats and their allies on the left received on average 52% of the two-block share, and the right of center parties an average of 48%.

In three election, in 1956, 1958 and 1973 the right of center block actually received more votes than the leftist block, but lost the election due to various technical reasons. While the fact that Al Gore won the popular vote has been covered extensively by the Swedish media, this historical fact about our own country is unknown by most Swedes.

In sum a sizable share of the Swedish public consistently opposed Social Democratic dominance, but with little electoral luck, and little influence, due to the "winner takes all" nature of the system.


52% of the vote and 85% of power is a pretty good deal, if you can get it.

Many Swedes consider the American system strange and undemocratic, but in my view this parliamentarian system where 50.1% of the vote gives you almost 100% of the power is in many ways less democratic than the American system that contains power sharing and protection of minority rights.



Wednesday, January 20, 2010

Sci-Fi Politics

In order to cheer up Democrats after yesterdays setback, I will simulate the 2052 presidential election.

Imagine a republican that does roughly as well as George Bush did in 2004, and receives:


58% of Non-Hispanic White vote

12% of African-Americans vote

40% of Hispanics vote

40% of Asians and others vote



In 2008 landscape that would have been enough for a bare majority of the vote.

But the population structure is changing, mainly due to immigration. The census bureau has
projections of the population structure. Now imagine that we are in the 2052 Presidential election, and the republican receives the same vote for each population group as Bush did. I also use census bureau data on voting patterns (they have the best data).

Let us assume that Hispanics and Asians close only half the gap in turnout with African Americans, as more become citizens and more of the citizens become politically engaged. (They don't close the gap entirely, because as this group will still be young and foreign born in 2052. African American turnout in 2008 was unusually high because of the Obama effect.)


Because of change of the demographic composition a Republican with the Bush 2004 winning share for each population group, enough for 50.3% in 2008, would get only
46.8% of the vote in 2052, close to McCain's losing share!

The median voter model predicts Republicans would somehow respond to this. But that model is overly simplistic, there is no guarantee that they will or even can. In Sweden between 1932-2010 Social Democrats were in power 85% of the time, in contrast to the simple prediction of the median voter theorem.

Unless Republicans sharply move to the left or the voters to right, today's result is just a temporary setback. Democrats only need to be a little more patient.

Monday, January 18, 2010

How to help Haiti


I have plotted remittances and aid for Haiti, as a share of GDP (the aid data is from the World Bank, and slightly different than the OECD data used in a previous post.)

In 2008 remittances were 18.7% of Haiti's GDP. In comparison foreign aid before the catastrophe was 10.4% of GDP. Agriculture is 28% of Haiti's GDP. These 3 sources of income are a combined 57% of Haiti's GDP.

They have one thing in common: they have a low (but not zero) marginal product of labor. Having more people will not make these sources of income go up proportionally.

Why would remittances not be affected by the number of people? The reason is that other countries limit Haitian immigration. According to Gallop about half of Haiti's population would emigrate if it could, one of the highest in the world. They have not emigrated because other countries limit how many Haitians they want to absorb.

Foreign aid is somewhat responsive to population, but far from proportional. Small poor countries tend to get more aid per capita.

More workers in agriculture certainly have some value, but again given how many people there are already in agriculture (two thirds of the labor force) more workers will not add much value. The scarce land limits the value of more labor.

Haiti's per capita GDP would be 43% higher if today's aid and remittances were shared on their 1960 population instead of their current population (assuming it aid and remittances were same, but ignoring agriculture).

The idea of population control has been discredited in rich countries. The reason is that with our institutions, wealth is produced by labor. More mouths also mean more hands, so per capita income is generally not related to population size. Since more people produce more ideas, growth may even increase with population size for rich countries.

Furthermore, people like Paul Ehrlich helped damaged the reputation of population control through alarmist predictions that did not pan out (because for the world as a whole they were not true). But as is common in intellectual debates there has been an overreaction, with many economists now dismissing population entirely as a problem.

While Ehrlich was wrong in general, population growth harms the standard of living in some specific cases, namely countries where a large share of income is derived from fixed assets (in Haiti's case land, relatives abroad and the generosity of foreign nations). The most obvious case where more people reduces the average standard of living is Saudi Arabia.

Economists should be more flexible in their analysis of countries problems. Population control in not an issue in nations where there are good enough institutions to have a manufacturing sector, but a huge problem where fixed assets are an important source of wealth, such as Haiti.



Saturday, January 16, 2010

Haiti's Economy

1. The worlds perhaps leading expert of Haiti's economy is a Swedish economics professor, Mats Lundahl. His research has emphasized the political economy of Haiti, overpopulation and a vicious cycle in resource use (which he links to ill functioning institutions). In "The Root of Haitian Underdevelopment" (1985) Lundahl writes:


Population growth in a country where the institutional environment did not enable an industrial sector led to more and more labor intensive agricultural techniques and products (food instead of coffee). This led to land erosion and a vicious cycle : erosion made useful land even more scarce, which leads to an even higher labor to land ratio, more labor intensive agriculture, more erosion etc.

Another problem that Lundahl studies in depth has been the kleptocratic nature of Haiti's political leadership. People who are interested about Haiti should read some of his work.


2. Between 1960 and 2008 Haiti received $8.6 billion (real) dollars in Official Development Aid, not a shocking amount by most comparisons. If someone would have invested this aid money in a bank account earning 3.5% real returns, it would be $16 billion, twice Haiti's nominal GDP. Haiti has become somewhat more aid dependent in recent year, as this graph shows.


The failure of development aid is no argument against catastrophe aid, quite the opposite. The problem with development aid is that it does not increase productive investment, it typically gets consumed. However the purpose of catastrophe aid is to get consumed. The lesson of 50 years of aid failure is that the west should give more catastrophe aid and less (no) development aid.


3. Much of the discussion about Haitis poverty centers around pre-WWII history. Any debate should however take into account that Haiti in 1950 was not poorer than the Dominican Republic. Only by the mid 1960s does Dominican start to diverge. if historical forces are at work they are operating with a lag (needless to say these figures are nowhere as reliable as for industrial countries).


4. There are about 800.000 Haitian in the US. If they were treated as their own a country their per capita GDP would be $28.200. The total income of this group is twice the GDP of Haiti (population 10 million).



Google Analytics Alternative