« Measuring sports performance | Main | What if all books were in Google? »
Why I believe David Card's results on immigration and wages
As MR readers will know, a famous David Card paper shows that the presence of many immigrants in a city does not much lower wages, if at all.
The obvious rejoinder is that cities with growing wages might attract more immigrants. The new immigrants will cause local wages to fall back down, resulting in a lack of regional correlation between wages and immigration. Without the immigrants maybe those locales would have had higher wages. So what does the Card paper really show?
Keep three points in mind:
1. The skeptical story does not consider all possible adjustments. Had there been no new immigrants in the growing cities, American workers would have moved in to take advantage of the higher wages, thereby pushing wages back down again. The U.S. has the most mobile labor supply of any developed country. So the net wage depression effect of new immigrants -- taking all supply adjustments into account -- still would be zero or very small for many places.
Note that this point, while it responds to Card skeptics, also creates some problems for the Card paper. It suggests that the labor market is defined at the level of the nation as a whole rather than the city or region. It is then no surprise -- no matter what your view of immigration -- if local labor conditions do not much correlate with local wage levels.
But invoking this point about the national scope of the labor market also blunts fears about how immigrants depress wages. Suddenly the question is not how many Mexicans are pouring into Texas, California, and Arizona. The question is rather how many Mexicans are pouring into the United States. But the larger the relevant market size, the better a job we will do absorbing immigrants. And the smaller the effect on domestic wages we should expect. To run a contrasting thought experiment, just try putting them all in Geneva, or better yet Soglio.
2. Have I mentioned capital mobility? Foreign capital flows into the U.S. all the time. Even when the Chinese buy T-bills, this frees up domestic private capital to put more people to work. Having more immigrants encourages more capital to flow in. If both labor and capital enter the country, there is no reason to expect immigration to lower domestic wages.
3. Are rising wage levels the key factor in drawing Mexicans to a region? El Paso appears to be one counterexample. In many cases proximity to the border and clustering effects seem to be more important. In that case the Card test has less of a problem with endogeneity. Still, I do not know the formal evidence on this point, so in the comments please pass along your expertise...
Posted by Tyler Cowen on May 22, 2006 at 06:52 AM in Economics | Permalink
Comments
Trade economists focus on the Rybczynski effect as another channel through which increased supplies of unskilled labor can be absorbed without affecting relative wages. Simply, a rise in unskilled labor supply leads to a more than proportional increase in the output of unskilled labor intensive industries. The result is an increase unskilled labor demand that offsets the supply increase, leaving wages unchanged.
Posted by: David Hummels at May 22, 2006 9:57:25 AM
Having gradually come into the Peak Oil camp,these arguments about
the marginal effects of immigration seem absurd. The point is
population, we need to aim for less than half of the present
population for the good of the grandchildren. How does immigration help
that? As Rep. Bartlett asks: Will your grandkids spit on your grave?
JSK
Posted by: JSKELLEY at May 22, 2006 11:10:18 AM
All this stuff is acadimic fluff. If you can't turn your bright ideas into useful political messages then it will never be tried out.
"Profiles in (economic) courage" would be a good working title.
Posted by: Huggy at May 22, 2006 11:59:25 AM
Flexible capital markets let America keep ratios of financial capital to labor high, but they don't maintain ratios of high human capital to low human capital people. Similarly for social capital, when 60% of Mexican immigrants support increasing the size of government vs 25% of native-born Americans, the institutional environment that is so essential for US productivity is endangered.
Posted by: Anon at May 22, 2006 12:25:14 PM
“Without the immigrants maybe those locales would have had higher wages. So what does the Card paper really show?”
“1. The skeptical story does not consider all possible adjustments. Had there been no new immigrants in the growing cities, American workers would have moved in to take advantage of the higher wages, thereby pushing wages back down again.”
But American workers would have moved out of other area to take advantage of higher wages in the city, which would drive up wages in the area they have moved out from!
“2. Have I mentioned capital mobility? Foreign capital flows into the U.S. all the time. Even when the Chinese buy T-bills, this frees up domestic private capital to put more people to work. Having more immigrants encourages more capital to flow in.”
What increases capital investment is not increased immigrants but increased savings. In addition, do companies really make a decision to increase the amount of capital investments based on an increase in illegal aliens or immigration in general?! Certainly, if one is looking to invest where there is excess labor and a paucity of capital one could do better than invest in the US. Do you have any evidence supporting your claim that immigrants encourage increased capital investments?
“If both labor and capital enter the country, there is no reason to expect immigration to lower domestic wages.”
Maybe, depending upon the amount of capital and labor entering the country, but wages would be higher if less labor entered the country increasing the capital per worker ratio.
Your two points in particular do show that wages would be higher with less illegal immigration, which is another wage of saying that illegal aliens lower wages for US workers! Thanks for the information Tyler.
Posted by: tom at May 22, 2006 12:26:35 PM
You can argue Rybczynski all you want to, but none of you are taking into consideration the considerable cost thast these illegals put onto our health care system, our higher insurance premiums, and the fact that over 30% of our prison population is made up of illegal immigrants. These costs MUST be weighed against the benefits for production and owners capital.
Posted by: jtapp at May 22, 2006 12:42:46 PM
Card is doing the usual economist thing -- "assume a can opener" -- as the old joke has it. Card has no access to the alternative world where immigrants haven't flooded into America. One of the effect of the flood in California is that millions of native born Americans have exited the state i.e. there is huge net native born emmigration from the Golden State. It should also be considered that a majority of the black population between 20 and 30 has simply left the job market.
The upshot is that there are no "closed economies" in the American labor market -- assuming a "closed economy" being another classic economist joke, one that Card can't help but tell.
Posted by: PrestoPundit at May 22, 2006 12:55:06 PM
Tyler,
There is no question that your arguments are correct IF and ONLY IF the skill and character quality of the immigrants is high (as in Chinese, Vietnamese, and other Asian immigrants). I do not think this is going to be the case for lower-skilled immigrants who, in the past, have shown little capability of moving up the congitive/economic ladder. Compare the Silicon Valley to Southern California. There is no question that much of Silicon Valley (and the techno-industries that have been spawned from it) would not exist if it had not been for the presence of the Chinese and Indian techno-entreprenuers that made it into reality.
Do you believe that the current and future wave of Latino immigrants will created the new biotech and nanotech "valleys" yet to come? Their record of academic achievement suggests that this is not likely.
The issue is not that ALL immigration is bad. Certainly immigration can and is good for the techno-economy. The issue is the QUALITY of the immigrants we are getting. We want immigrants that will create the next biotech and nanotech valleys, not a new underclass. It is the refusal of the pro-immigration side to consider the QUALITY of the people coming in that obfuscates the immigration issue.
PC liberalism will be the death of the West if we are not careful.
Posted by: Kurt at May 22, 2006 12:57:20 PM
The African-Americans who are being fired to make way for cheaper illegals don't care much about what economists have to say, and they don't care much about the national labor market.
Posted by: save_the_rustbelt at May 22, 2006 1:17:20 PM
http://www.fordschool.umich.edu/rsie/workingpapers/Papers426-450/r448.pdf
"Recent literature on the labor-market effects of U.S. immigration tends to find little correlation between regional immigrant inflows and changes in relative regional wages. In this paper we examine whether immigration, or endowment shocks more generally, altered U.S. regional output mixes as predicted by the Rybczynski Theorem of Heckscher-Ohlin (HO) trade theory. This theorem describes how regions can absorb endowment shocks via changes in output mix without any changes in relative regional factor prices. Treating U.S. states as HO regions, we search for evidence of regional output-mix effects using a new data set that combines state endowments, outputs, and employment in 1980 and 1990. We have two main findings. First, state output-mix changes broadly match state endowment changes. Second, variation in state unit factor requirements is consistent with relative factor-price equalization (FPE) across states, which is a sufficient condition for our output-mix hypothesis to hold. Overall, these findings suggest that states absorb regional endowment shocks through mechanisms other than changes in relative regional factor prices."
Posted by: Mr. Econotarian at May 22, 2006 1:21:42 PM
Comments have started to get really embarassing - maybe it is time to reconsider policy given the heat/light ratio generated in response to immigration posts (which have been your most interesting ones of late)?
At least Sailer hasn't come by yet and claimed that Card's study doesn't consider cost of living differences. That would be REALLY embarassing for the future of quality comments ...
Posted by: Matt at May 22, 2006 2:25:39 PM
1. I cannot understand how Cowen can write a post with the title “Why I believe David Card's results on immigration and wages” but in the text admit that he believes the US is better seen as an integrated labour market. This is the main critique of Card!!!
Yes, the US is better seen as a national labour market. This is what Harvard Borjas pointed out, and this is where the 7-8% reduction of low-skilled wages comes from. Borjas estimates 3-4% reduction in the wages of the lowskilled if you expand the supply by 10%.
http://ksghome.harvard.edu/~GBorjas/Papers/QJE2003.pdf
http://ksghome.harvard.edu/~GBorjas/Papers/cis504.pdf
Borjas is the most important immigration economist in the US, and his research represent the forefront in the field. I don’t understand how Cowen can keep ignoring his results and referring to Card, when he in the text agrees with Borjas critique of Card (the US is a national market, regional effects are secondary).
“But invoking this point about the national scope of the labor market also blunts fears about how immigrants depress wages. Suddenly the question is not how many Mexicans are pouring into Texas, California, and Arizona. The question is rather how many Mexicans are pouring into the United States. “
Again we see a straw-man argument by Cowen.
Nobody in the threads here or in the immigration-skeptics in general has claimed the regional effects in the labour market are most important (this is of course distinct from say California tax burden effect). But the national effects are very significant.
There are about 16 million people in the workforce nationally with no high-school diploma, of course importing 7 million additional Mexicans will impact their wages, even nationally.
Succinctly: If you believe the US should be seen as one market you tend to be in the “immigration hurts lowskill income” camp, if you think we can look city by city you are in the Card “supply does not matter” camp. You can’t be in both camps at once.
2. “Having more immigrants encourages more capital to flow in”
Low skill labour and capital are in many cases better seen as substitutes, rather than complements. At any case even if there is more capital there will hardly be more capital PER WORKER, which is what determines wages.
3. Sorry, but irrelevant of Rybczynski-style output change the theorem has little relevance here in terms of factor prices (wages). If you believe this framework you should believe in factor prize equalization, so that wages of low skilled in Mexico and the US are the same!
Clearly we are in a type of corner solution in this regard, furthermore almost all the Mexicans and low-skill labour work in the non-tradable sector. This makes sense; the low skill labour intensive traded sectors are competed away by china, whereas the low skill can still earn good money in the US by working in the non tradable sector. They earn OK because there are few of them.
4. As I wrote earlier the 10% or so reduction in low-skill Americans earnings is not the biggest problem, even if denying it is a bad sign. It is the effects that go though other channels than the labour market:
• Costs though welfare spending, each immigrant with less than high school degree represents a nett present value cost of 89.000 $ (1997 dollars). 20 million additional Mexicans with no high school diploma, a reasonable assumption given the Senates proposal, means about 2 trillion $ in costs to US taxpayers.
• Increased crime, the expected increase murder rate alone is equitant to US yearly losses in Iraq (but unlike the Iraq war the victims will not be white middle class, most will be other Hispanics, blacks or poor)
The massive political change. Anon remembers incorrectly, the share that want to expand government of whites was 35% of whites vs. 60%, for Hispanics (holding almost constant in second generation Latinos). According to Pew Hispanics whites were 220% as likely as Hispanics to be “enterprisers”, hardcore anti-government conservatives. Even the minority of republican leaning Hispanic tend to be big government Conservatives, anti-gay and ant-abortion, but pro government.
In 30-40 years almost half of US voters will be part of the underclass, leading to drastically stronger demands for taxation and redistribution, and move towards Welfare state economics. The disastrous economical impacts of this political change will make everything else second order.
Posted by: Teller at May 22, 2006 2:42:40 PM
The most realistic model would fall in between Card's purely local one and Borjas's national one: without foreign immigration, American labor would flow from low wage to high wage cities, but of course it wouldn't be a friction-free flow. American workers would require a wage premium to get them to move away from family and friends in their hometowns and off to some boomtown in the desert. But that wage premium gets hammered down by illegal immigration, so lots of Americans don't find it worthwhile to move from their Rust Belt homes.
In contrast, in the 1950s and 1960s, moving to Southern California was almost a no-brainer decision for millions of average American Joes -- the LA economy was booming, wages were relatively high because of immigration restrictions, and the cost of living was relatively low. There was a sizable increase in the welfare of the average American because Southern California was open to them. But, that's no longer true.
From the Mexican illegal immigrant's point of view, he has similar wishes -- he too doesn't want to move that far in distance, or in climate, and would like to live somewhere close to other Mexican immigrants. But, for him, the wage premium of sneaking across the border is so huge that he has already made up his mind to move, so of course he will go to the boom towns like Las Vegas rather than to the bust towns like Pittsburgh.
The other thing to keep in mind is that most of the American boomtowns in recent decades have been west or south of the Rust Belt -- i.e., in the general direction of Mexico. This means that for Mexican illegal immigrants, there is less of an adjustment necessary in terms of climate, distance from home, and local demographics than if this process wasn't going on. To help you understand this better, imagine that Quebec was the main source of illegal immigration. Then, the American boom towns would be located away from them. But, Mexicans are lucky in that the boom towns are coming towards them.
Posted by: Steve Sailer at May 22, 2006 3:30:44 PM
IMO, a little-discussed but profound issue with illegal immigrants is illegal employers. The assumption behind all this economic discussion is that "a job is a job". The problem is that many if not most illegals work for illegal employers, meaning those who pay under the table and don't bother with taxes, unemployment insurance, workplace safety regs, etc. In my opinion, this is the biggest reason immigrants "do jobs Americans don't want to do" - Americans won't, in general, take jobs under the table.
If you are a marginal employer, you _want_ illegals since you can pay them under the table and not worry about them blowing the whistle on you about unsafe working conditions, no employer SS, etc.
There are so many illegal employers that whole sectors of the economy are defined by them, so it is difficult to operate a legal business even if you wanted to in those sectors. And, the fly in the ointment for those who advocate "enforcement": I'd guess that most of these employers are small mom&pop operations which will be monstrously expensive to track down and deal with.
Posted by: Foobarista at May 22, 2006 3:42:02 PM
More on the "illegal jobs" - not all the aspects of an illegal job are necessarily bad for the employee; often part of a restaurant worker's pay is room and board, with "board" meaning a bed in a crowded apartment near the restaurant. It may suck, but low-pay jobs suck generally and at least a diligent saver can save quite a lot of money working this way. So part of the answer may be in making it easier to do these sorts of arrangements legally without a ton of regulations and extra taxes getting in the way.
Also, in Silicon Valley at least, even illegal workers are usually paid more than the minimum wage, so it isn't necessarily wage laws that drives the work underground.
Posted by: Foobarista at May 22, 2006 3:53:12 PM
Borjas and Katz get 8% lower wages only by assuming zero shift in capital investment, industry composition, etc. in the face of immigration. With those relaxed, the effect drops to negligible levels.
You don't need to use Card's analysis to solve the endogeneity problem. The Ottoviani and Peri paper on complementarities between immigrants and natives (cited in a previous post on MR) used a very clever instrumental variable approach. They pointed out that immigrants tend to move to concentrations of their own ethnicity, holding constant all other factors. Then they used the preexisting ethnic concentrations across cities (before the immigration flow they were studying occurred) as an instrument for actual immigration. The results were basically unchanged.
Posted by: srp at May 22, 2006 3:59:35 PM
Even this novelist understand the basic principle here.
http://www.washingtonpost.com/wp-dyn/content/article/2006/05/19/AR2006051901524.html
"To assert as some have that illegal immigrants do not depress wages because they do the jobs Americans refuse is the kind of nonsense economists speak when they strain to be counterintuitive. It is similar to saying that cheap imports do not hold down prices. If employers paid higher than substandard wages, Americans (who famously do almost anything for money, including eating worms, shooting themselves from cannons and listening to Barbra Streisand sing) would take these jobs."
Posted by: Econonut at May 22, 2006 4:06:11 PM
We can dispense with the 8% (7.4%?) figure of Borjas because it neither captures output expansion nor changes in capital flows, both of which mitigate the effects - Borjas acknowledges the true number is closer to 1-2%. If we then accept that there are no relevant regional differences, which Teller seems ready to do, I think Tyler has won his battle. No straw man there ...
I was under the impression that there is consensus on national cost-benefit estimates: net benefit for immigration - perhaps some are only looking at the effect on fiscal health? A similar (mis-)emphasis shows that smoking helps the country by decreasing social security payouts later in life, when obviously the effect on the economy is negative.
Re: the criminal tendencies of hispanic immigrants see Cowen's older post http://www.marginalrevolution.com/marginalrevolution/2006/03/does_mexican_im.html#comments.
Also, it seems Teller is either defining "under-class" up or is very, very pessimistic about the future of the USA when he/she claims that almost half of our population will comprise it. This is utterly ridiculous ...
Posted by: Matt at May 22, 2006 4:21:49 PM
You guys are hilarious!
1. The Ottoviani and Peri paper strongly CONFIRMS our arguments, and REFUTES yours and Cards!
What they do is to include capital, and also to have less of substitution between foreign and domestic workers. (if American unskilled and Mexican unskilled are less of substitutes the later will effect the wage of the former less, fair enough).
What do they find for unskilled? Gee, it seems the own elasticity is very similar to Evil racist Borjas, 0.3. On what basis do you call that “negligible”? Maybe it is to you, but not to poor Americans.
The effect on unskilled Americans in the 1990-2000 years alone was a drop in wages of 4.5% (other specification give even stronger results). Borjas get’s 7.2% dropp 1980-2000,. These are very similar figures, and both suggests that common sense and economic science is correct: Supply effects wages. Amazing no?
Now comes the funny part. Do you know why the effect on the low-skilled Americans have not been even stronger? Because immigration also includes High skill foreigners. In the long run the composition of the labor force is what determines relative wages for different groups, high skill workers pull up the wages of low skill, wheras the low-skill immigrations press downward the wages of low-skill, and help high skill.
So given the article you yourself cite, what do you think is smarter: the present policy of dramatically INCREASING extremely unskilled immigrants from Mexico, or instead doing what I, David Victor Hansen and Sailer propose, instead taking in more elite immigrants from Asian and Europe?
Economics is a science, it is not some sort of religion where you can refer to the tone of language used in the abstract and assume it gives your fait support.
Claiming Tyler has “won the battle” by referring to a paper that widens the scope, includes capital, reduces substitutability and still gets the exact opposite results as Card and Tyler are claiming is really completely pathetic.
2. If you want to understand how low economist has sunk just look at how NYT and Cowen explain away Hispanic crime:
controlling for background characteristics Hispanics commit fewer crimes (than blacks in practice). Not controlling they commit 3-4 times as much as whites, and are put in prison 3-4 times as much.
What this only means is that their extremely high crime rate can be explained by their traits (uneducated, low-income, broken homes)
Now here is a question from Econ 101:
Does explaining an effect by referring to background characteristic remove the effect?!? Are you stupid? As I wrote, when you are importing Mexicans you are importing their background characteristics, in this case leading to massively higher crime.
3. The endogenously effect has nothing to do with the US being one largely integrated labor market, which is also the result of this paper, in direct contradiction to Card.
(Furthermore this “instrument” only shifts back the question. Immigrants didn’t magically grow out of the ground in California, Nevada and Florida, the first groups moved to the fast growing sun belt, not to the coal country in West Virginia., not that it matters much)
4. By 2050 some 40% of the US population will be African American and Hispanics. If you don’t consider these groups underclass than you are a very optimistic person. (of course even today not all blacks and Mexicans are underclass, but they vote with the extremely high percentage of their ethnic group that are)
Posted by: Teller at May 22, 2006 5:00:54 PM
In other words, the law of supply and demand has been repealed?
Posted by: Hans Gruber at May 22, 2006 5:12:17 PM
Teller I think you are confusing posts. My point was take Borjas' estimate and adjust it for expansion of output in drop-out intensive industries and capital allocation and you get something closer to 1-2%, a point Borjas himself admits.
As for criminality, it can go both ways. It depends on who is being displaced from urban centers and what the effect this has on their rate of criminal behavior. The entire point is that Sampson claims the net effect is positive. You can't just look at the static rate on the group in isolation.
And no you don't have to be an optimist to refuse to count all or even an "extremely high percentage of" blacks and Latinos as the underclass. And since I vote along with them pretty frequently, I don't really accept that as an appropriate metric either.
Posted by: Matt at May 22, 2006 5:18:36 PM
I always find it interesting how so many economists attempt to justify the law of supply and demand for everything EXCEPT labor. On that subject, they will strain themselves to the n-th degree to justify a counterintuitive position.
Posted by: mike at May 22, 2006 5:44:14 PM
I also think it is hilarious, Teller, that you consider the Ottaviano paper to be on your side of this debate, considering they would never agree with that assessment. Read their paper - unskilled immigration has increased average wages, and the effect on native high school dropouts is less than 2% under their preferred estimates.
Posted by: Matt at May 22, 2006 5:45:48 PM
No matt you are confusing the argument. For the 18th time, we are discussing the Marginal effect of Mexican immigration I nteh future, NOT the average effect of all immigration in the past.
The reason low-skilled have not done worse is not mainly capital expansion or industry mix, it is the fact that overall immigration has been balanced in these years. Even the estimates in the paper that includes lower substitution and fully elastic capital gets strong effect for low-skill wages if you expand the numbers. Just read the paper.
Card claimed there were no effects, by only looking at one city or industry at a time. Both Borjas and Ottoviani and Peri soundly refute this idea. Cowen is silly when he first writes in the headline that Card is right, than in the text comes to his senses and acknowledges the US is one labor market.
But as has become common with pro-immigration groups the way you present data to your readers and the actual data themselves are not always the same. Take crime for instance, where there is no way no how you can deny the overrepresentation of Hispanics. Saying they are
young/men/uneducated/poor/from broken homes
does NOT change the overall increase in crime, it only explains it.
Now look what Cowen does:
He ideologically want to argue that Mexicans don’t increase crime. The best he can do is to link to a paper that shows they are imprisoned twice (1.983 times) as much as Americans and thus 4 times as much as whites!
“Masking the divide” got 3.7 times as much crime as whites, and 1.7 times the national average for all Hispanics. http://66.165.94.98/stories/maskdiv0501.pdf
This is the raw actual imprisonment (remember a wast literature shows that incarsaration and victims reported crime are almost identical, so don’t come with the racist judicial system excuse).
Hispanics commit much much more crime than whites, and significantly more crime than the national average.
This is how I made my calculation of a murder rate of lower bound 8.5 per 100.000, and
Cowens source even more strongly support my argument, giving a rough 10 per 100.000 murder rate, I seem to have underestimated how crime-prone Mexicans are. Thanks Tyler.
This gives 500 EXTRA murders for each 10 million Mexican immigrants, so 1000 more for 20 million, compared to 7-800 Americans killed in Iraq.
Sampson is a complete idiot, pure and simple. What he does has nothing do to with “who is being displaced from urban centers and what the effect this has on their rate of criminal behavior”.
He just fools readers who don’t pay attention or don’t understand statistics. He ADJUST(!!!) for family and neighborhood background to get his “result” that Mexicans commit less crime. That means he is EXPLAINING the high crime rate through being underclass, and than REMOVES this effect from the data before presenting it to readers.
Do you know what that means? It means lying. Do you mean what it means to report or believe it, if you understand simple econometrics?
“since I vote along with them pretty frequently”
That explains a lot doesn’t it.
Posted by: Teller at May 22, 2006 5:46:18 PM
For a quick refutation of the NYT times op-ed by Sampson supposedly refuting the notion that illegal immigration raises crime rates, see:
http://www.vdare.com/Sailer/060312_sampson.htm
Posted by: Steve Sailer at May 22, 2006 5:54:35 PM
Well regardless of what Teller thinks of my voting preference (Democrat), I am not what a part of the underclass so I think the point holds.
And I understand econometrics just fine. I will read over the Sampson stuff with your points in mind, but it is tangential to the point about wages. And no one here is race-baiting as far as I can tell so I'd like to tone this whole set of posts down a notch or two.
And this bears repeating, that although Ottaviani and Peri have the same (unadjusted) elasticity of relative subsitution, the additional effects they note back up Card's observation that wages haven't decreased much, if at all, for high school drop-outs.
Posted by: Matt at May 22, 2006 5:59:37 PM
Matt my comment on your voting preference refers to prefrenses for expanding government and strengthening the left by increasing the underclass, not you being underclass personally.
Some examples of logical problems on your side:
· I say there will be massive costs of the using going to the left because of the explosion of the two ethic groups that compromise the pro-welfare state underclass. You say not all blacks and Hispanics are underclass. I reply that maybe not, but then tend to vote with their ethnic groups, even those who do well economically.
You responded that “no, I am a Democrat, and I am not underclass”.
A gives B does is hardly refuted by saying that “I am B, but not A”.
* They get -4.5% by including capital in estimates otherwise similar to Borjas. So clearly capital will not change this. (They get their “preferred” estimate of 2.2% by putting in parameter values they assume themselves. The ones with estimated ballpark elasticities is somewhere in the middle).
Even with their personally chosen parameter values they get a strong effect of wage elasticity of low-skilled to the increase of the share of low-skilled immigrants in the economy, in calculations that assume the country is a integrated labor market.
So here is the difference:
Card claims there are small effects from immigration on wages. He does this by looking at single cities. Tyler writes that no problem, let’s take in more unskilled Mexican, there are small effects.
Borjas AND Gianmarco and Peri don’t accept looking at single cities. They both look at the US nationally, the latter using capital and imperfect substitutability of foreign born to reduce the effect the former estimated. Yet BOTH get higher than Cards elasticities for low skill.
How does this prove Tyler right exactly?
It is the elasticity that is the key, not the historical experience of wages. Why? Because we are specifically discussing low skill immigration from Mexico (70% no high school degree, 4% college). The historical experience was one of bringing in both unskilled and high skilled. While the paper you refer to estimates a very strong downward pressure on unskilled American wages from unskilled forigners it also includes the upward pressure from high skill immigrants.
The last point is NOT relevant to this discussion, or actually it further strengthens the point. If you believe Gianmarco and Peri the US should bring in less Mexicans and more Europeans/Asians, not vice versa.
So when you claims “the additional effects they note back up Card's observation that wages haven't decreased much, if at all, for high school drop-outs.”
this is technically true because of the historical experience, but compactly irrelevant, since they also estimate elasticities that suggest Borjas is right to say unskilled Americans wages are sensitive to the number (really share) of unskilled workers. Card however lives in the liberal world of zero elasticities, nothing effects anything.
It is not the “that wages haven't decreased” that matters most, it is how much WILL they change of the proposed policy of massive increase of unskilled immigration.
The question of crime is not “tangential”, because you have decided to make the discussion only about wages, whereas a good economist would consider all economic effects of immigration.
I personally think the effect of wages is there and become angry at people who lie and say it is not, but I also think it is secondary. It is possible to compensate the poor Americans for the lowering of their wages if you have a net increase, even if it is expensive.
But the additional effects of immigration, in welfare subsidies, crime and political distortions are massive.
The 1-2 estimated effect on OVERALL wages (that I have never denied will go up due to unskilled immigration) is less even than the DIRECT estimated welfare costs (equal to 3-4 trillion in economical costs, if you include the distoritionary costs of taxation).
Posted by: Teller at May 22, 2006 6:35:36 PM
PS. The 1-2% refers to the estimated increase in all american wages by the pro-immigration Ottaviano/Peri paper
http://www.nber.org/papers/w11672.pdf
They get minus 2.4% unskilled wages 1990-2000 alone even with the relativly low estimated substitution, -4.5% by assuming low-skill forigners directly compete with americans with the same skill.
If what I wrote is confusing very short:
Anti immigration Borjas and pro-immigration Ottaviano/Pri both disagree with Card, and say unskilled americans wages are sensetive to the number of unskilled forigners. However Ottaviano/Pri agree with card that the pressure downward the last decade was not very strong, because the number of educated forigners pushed up low-skill american wages.
In regards to the marginal political question of Mexican immigration the elasticity is relevant, not the historical experience or mixed immigration.
Posted by: Teller at May 22, 2006 6:43:24 PM
Ok I'm prepared to declare victory and move on. While granting none of your objections to my points, I am happy that you are prepared to acknowledge a rise in overall average wages and a smaller (by half) decrease in wages to native high school drop outs than Borjas' headline number. The gap between us is shrinking!
Since this was a discussion about wages and not crime, I'm not going to rush to familiarize myself with Sampson's work in order to agree or disagree with you about that.
And I think it was fair (if unimportant overall) for me to assume that you were claiming that '40% of the nation is going to become a part of the underclass', not that '40% of the nation will be of an ethnic group that historically votes for the liberal party that wants to increase the welfare state'. Since I assumed incorrectly, I apologize.
Posted by: Matt at May 22, 2006 6:53:58 PM
Also, -2.4% is the result of their "high" estimate, not median or low. Depending on whether subsitutability is allowed to vary between education groups (one of their broader points) the median estimate for high school drop outs is either -1.7 or -1.2%. For all wages it is 1.1 and 1.2%.
http://www.econ.ucdavis.edu/faculty/gperi/Papers/perott_aggregate_6.pdf
(Non-paying link)
Posted by: Matt at May 22, 2006 7:02:39 PM
A quick way to get a sense of the impact of immigration on crime is to look at the hundreds of pictures on the Los Angles Police Department's Most Wanted list:
http://www.lapdonline.org/all_most_wanted
Make sure to check out Vanessa Lanza Etorneau and Corrina Kowalski, while you are at it.
Posted by: Steve Sailer at May 22, 2006 7:05:35 PM
1. You keep saying this is a discussion about wages. No, it is a discussion about the economical costs and benefits of Mexican immigration, including wages, and also including crime, taxes, assimilation, welfare expenditure and intergenerational mobillity. All economic effect matter. The debate started because Tyler Cowen wrote:
"David Card and others have plenty of data on how well the second and third generations of Latinos do in assimilating and entering the mainstream of American life. I find the overall portrait a reassuring one. I will look for data on Mexicans per se and let you all know if I find anything useful."
So far the effect has been, I think even most pro-immigration advocates agree, far from reassuring.
Secondly even if you only want to look at wages what matters in this sub-thread
* Is David Card right about wage elasticity and looking at single cities instead of the US?
* Is the expected future wage effects of massivly increasing lowskilled Mexican immigration on unskilled Americans negligible?
In both cases even if one accepts everything in the two (somewhat conflicting) Ottaviano/Peri papers you refer to the answer is NO!
2. Actually 2.4% is not the highest estimate, they have 1990-2000
-7% (fixed capital)
-5.5% (fixed capital, foreign workers perfect substitutes)
-0.2 to –2.4% with different substitutability parameters.
Note also that we are looking at two different version, they have lowered some of their parameter values for substitutability in the one you are citing. I don’t quite understand why in the previous version the perfect capital elasticity and substitute gave -4.5%, and in the later version -5.5%. What assumption did they change?
(I am also curious what their estimates are 1980-2000 like Borjas, instead of only 1990-2000)
2.You “declare victory” because I repeat something I wrote already a week ago? That wages go up, but not by far enough to compensate all the other effects? The only point that I am perfectly willing to grant is that Borjas overestimates the historical experience by not including Capital, although the effect may not be as straightforward as you think if capital is also a substitute for unskilled labour.
In the Borjas-Card debate of ELASTICITY, the question that matters most for future effects on wages, the Ottaviano/Peri work only strengthens Borjas.
3. You write: “I am happy that you are prepared to acknowledge a rise in overall average wages”
I wrote May 16:
“I do not doubt that there are direct economic gains from immigration, throu
Posted by: Teller at May 22, 2006 8:18:10 PM
1. You keep saying this is a discussion about wages. No, it is a discussion about the economical costs and benefits of Mexican immigration, including wages, and also including crime, taxes, assimilation, welfare expenditure and intergenerational mobillity. All economic effect matter. The debate started because Tyler Cowen wrote:
"David Card and others have plenty of data on how well the second and third generations of Latinos do in assimilating and entering the mainstream of American life. I find the overall portrait a reassuring one. I will look for data on Mexicans per se and let you all know if I find anything useful."
So far the effect has been, I think even most pro-immigration advocates agree, far from reassuring.
Secondly even if you only want to look at wages what matters in this sub-thread
* Is David Card right about wage elasticity and looking at single cities instead of the US?
* Is the expected future wage effects of massivly increasing lowskilled Mexican immigration on unskilled Americans negligible?
In both cases even if one accepts everything in the two (somewhat conflicting) Ottaviano/Peri papers you refer to the answer is NO!
2. Actually 2.4% is not the highest estimate, they have 1990-2000
-7% (fixed capital)
-5.5% (fixed capital, foreign workers perfect substitutes)
-0.2 to –2.4% with different substitutability parameters.
Note also that we are looking at two different version, they have lowered some of their parameter values for substitutability in the one you are citing. I don’t quite understand why in the previous version the perfect capital elasticity and substitute gave -4.5%, and in the later version -5.5%. What assumption did they change?
(I am also curious what their estimates are 1980-2000 like Borjas, instead of only 1990-2000)
2.You “declare victory” because I repeat something I wrote already a week ago? That wages go up, but not by far enough to compensate all the other effects? The only point that I am perfectly willing to grant is that Borjas overestimates the historical experience by not including Capital, although the effect may not be as straightforward as you think if capital is also a substitute for unskilled labour.
In the Borjas-Card debate of ELASTICITY, the question that matters most for future effects on wages, the Ottaviano/Peri work only strengthens Borjas.
3. You write: “I am happy that you are prepared to acknowledge a rise in overall average wages”
I wrote May 16:
“I do not doubt that there are direct economic gains from immigration, through lower prices for services for example.”
“[Keith] claim[s] that this data underestimates the economic gains, and point to the [Ottaviano/Peri] paper that estimates a (not very robust) average real wage of natives receives a positive effect between 1.1% and 1.2% from immigration in the period considered. But rather than overestimating the costs both this papers grossly underestimating the costs of unskilled immigration. Let me give you some examples:
· "Welfare dependency…"
· "Crime…"
· "Costs of worse economic policies….”
4. There is no need to “rush” to understand that Sampson, whom you, (through Cowen) used as factual support to is full of it. Just read his very short NYT article and notice the little words “adjust for family and neighborhood background”
I repeat, Cowens own data shows Mexicans are imprisoned 4 times as much as whites and twice as much as all Americans. Sounds like a decrease in crime to you? It is further hilarious that Sampson writing that third generation immigrants commit 28% MORE crime than second generation is used as argument for immigration?
http://www.wjh.harvard.edu/soc/faculty/sampson/articles/2006_NYT_OpenDoors.pdf
Sure sounds "reassuring" to me. (although this also seems to be part of his idiotic calculations that "adjust for background charactristics, I pray to god the pure total effect does not look like this)
5. Yes it was fair to make that assumption since you had not read the previous parts of the debate.
I am still waiting for someone, anyone, especially pro-immigration libertarians, to make any counterargument to my main point in this debate during the last week, that Expanding the ethnic groups that constitute the Underclass from 20-25 to 40-45% of the population will have dramatic negative effects on economic policy. Obviously a Democrat like Matt wants this to happen, they are importing voters. But why do pro-market people support moving the US towards a European Welfare state?
Posted by: Teller at May 22, 2006 8:18:18 PM
Haha, ok this argument is obviously part of a larger discussion re: the economics of immigration policy (and I have obviously not been paying attention to everything that Teller has written on this subject!) ... by no means was I playing "gotcha" when I said you were prepared to admit wages go up, I was just saying that you admit wages go up, which is 3/4 of the (narrow) argument about wages.
As a non-Democrat who nevertheless votes for them pretty frequently, I take issue with this statement: "Costs of worse economic policies…" but I'm aware that simply disagreeing isn't going to change your mind.
Regarding the debate about elasticity, remember that Card is only implying an elasticity from changes in wage ratios - this is not the same kind of elasticity, and in fact it is possible for all 3 papers to (roughly) agree. If Borjas is correct, and then the adjustments that O&P make are appropriate, then the result is something that looks like what Card found evidence for, i.e. very little change in wages for high school drop outs. No need to necessarily read them as complete opposite ends of the ideological spectrum.
[At the risk of opening a can of worms, we can make unambiguous assumptions about the change in crime rate under 2 scenarios: if a) illegal aliens have lower absolute crime rates then the crime rate will go down, and b) if illegal aliens have higher crime rates adjusted for income and demographics then the crime rate will go up. Under any other situation we may think we know the answer but it is not assured.
My (lack of) understanding of Sampson from Tyler's post is that since they fit in between, additional evidence (like lower crime in inner cities not attributable to incarceration rates or the decline of crack prices) led him to believe that the secondary effect of lower adjusted crime rates swamps the primary effect of higher overall rates. You might justifiably be skeptical of this claim, but it isn't ridiculous on its face.]
Posted by: Matt at May 22, 2006 8:40:54 PM
Teller,
Does the size and scope of government hand-outs in the U.S. mean that the term you used "pro-immigration libertarian" is acually an oxymoron or perhaps a contradiction in terms?
Posted by: jim at May 22, 2006 9:05:04 PM
Explain to me how higher taxes and more government redistribtuion will have no negative effect on the US economy. Again, are we pretending to live in the liberal alternative reality where there are no elasticities?
Card does estimat elasticites, and O&P even directly compare their estimate with Cards (that are close to zero).
Again, the weak effect in O&P is because of the skilled immigratin that accomponed the unskilled, they actually find pretty strong direct pressure from unskilled immigrant to unskilled wages.
There is no basis to “adjust” for the income or other background factors of Mexicans when considering crime. Maybe you can *explain* their high crime rate through lower income income and underclass caractristica, but since they HAVE these caractistica the high crime effect will be there.
That’s like saying Math Phd departments should not have GRE tests, because given the GRE tests everyone does as well.
Jim:
Good point, unless you principially object to the state and don’t care about real world outcome.
In practice it seems most libererians are just soft “hearted” and just prefer to ignore the facts on the issue of immigration. Also, as in many issues, American liberterians tend to replace analytical reasoning with identity. “biggoted southern Conservatives oppose open borders, these people are dumb, therefore open borders are good”.
Each unskilled immigrans nett costs the taxpayers about 100.000 2006 dollars in benefits. Since on the margin the cost of each tax dollar is likely to be 1.5-2 dollars from the private sector this translates to 150.000-200.000 dollars per unskilled immigrant.
For this reason the magnitude the of the wage effect also matters.
Posted by: Teller at May 22, 2006 11:40:49 PM
Tyler,
Have you read Borjas QJE 2003? That paper considers the labor market to be national. Borjas exploits the variation in immigrant supply shocks across different education-experience groupings (such as high school dropouts with 5 years experience, and high school dropouts with 20 years experience). He finds significant but small negative effects of immigration on the wages of directly competing natives, defined by the education-experience grouping.
Still have to think about the implications for the "capital adjusts" argument in light of the finding. However, Borjas is only using contemporaneous immigrant supply shocks.
Posted by: brian at May 22, 2006 11:56:20 PM
"Still have to think about the implications for the "capital adjusts" argument in light of the finding."
One thing that's clear is that illegal aliens are a substitute for investing in labor-saving technology, just as an ample supply of slaves was a major reason for the lack of technological progress in the Roman Empire.
For example, in much of the country, there are robot car washes that are open 24x7 because they don't use any workers. But in LA, I've never seen a robot car wash. All the car washes use swarms of illegal aliens. But they are only open about 8 hours a day, so good luck getting your car washed before noon.
Similarly, the great American tradition of mechanizing farm work has ground to a halt in California since illegal aliens broke the UFW union in the 1980s.
Posted by: Steve Sailer at May 23, 2006 1:02:32 AM
There are about 16 million people in the workforce nationally with no high-school diploma, of course importing 7 million additional Mexicans will impact their wages, even nationally.
Of course? That's mighty overconfident of you. If the economy grows briskly enough, the demand for the skills of these (now) 23 million people without high-school diplomas might well bid up their wages -- giving them increases in living standards. Of course, for such robust growth, an adequate supply of capital is needed. Mostly, getting this capital has not been problematic for the US economy, and mostly, real wages for American workers have tended to increase. American workers -- even those at the bottom of the wage latter -- were a lot better off in real terms in 1930 than they were fifty years previous, at the dawn of an immigration-induced workforce expansion significantly larger in scale than the one we're witnessing now.
Posted by: 99 at May 23, 2006 1:14:36 AM
Dear 99:
Ceteris paribus is the term economists use.
Posted by: Steve Sailer at May 23, 2006 1:18:55 AM
In contrast, in the 1950s and 1960s, moving to Southern California was almost a no-brainer decision for millions of average American Joes -- the LA economy was booming, wages were relatively high because of immigration restrictions, and the cost of living was relatively low.
Highly dubious stuff here.
Of course, moving to Southern California is no longer the "no-brainer" it might have been in 1956. Things change over a half century. Manhattan's problably no longer the no-brainer it was as a domestic migration destination in 1830, either.
And that's because Atlanta, Charlotte, San Antonio and Tampa are the no-brainers of 2006. Indeed these cities are attracting lots of newcomers, from both domestic and international sources.
If it's any consolation, as America gets more expensive, it will seem less attractive to foreigners, many of whom will have managed to narrow the wage gap between themselves and American workers.
Posted by: 99 at May 23, 2006 1:24:17 AM
Similarly, the great American tradition of mechanizing farm work has ground to a halt in California since illegal aliens broke the UFW union in the 1980s.
Really? I keep hearing about how American farmers are feeding more and more people. Are you saying American agricultural productivity is in decline? Damn DOA has been lying to us all these years, and those declining food budgets have been illusory.
Posted by: 99 at May 23, 2006 1:28:45 AM
Steve wrote:
"One thing that's clear is that illegal aliens are a substitute for investing in labor-saving technology, just as an ample supply of slaves was a major reason for the lack of technological progress in the Roman Empire."
The point about the Roman Empire is a contentious one that many economic historians do not accept but that's a whole other topic (in any case, slavery in the American South did not stop the cotton gin from being embraced by plantation owners). As for the general argument, you are confusing what economists call technology (ideas or techniques to combine labor and capital in more efficient ways) with what non-economists call technology (gadgets and machines that do cool things - or what economists call capital). The former is essential for economic growth but is also not affected by immigration (Intel does not stop producing microchips just because there are more Mexicans in Santa Clara). Adoption of labor-saving devices (capital accumulation) will be affected by immigration but that's not what drives economic growth. The fact that immigrants wash your car instead of some $100,000+ machine does not, alone, affect the living standards of Americans.
Second of all, capital could be a complement rather than a substitute for labor in some sectors of the economy. In this case, a greater labor supply will make investment more profitable. I think some make this argument in regards to East Asia and its fertility bubble being a proximate cause of rapid capital accumulation and economic growth, but I'm not familiar with that literature.
If you want to show immigration is bad, stick to the traditional story: unskilled immigrants cost society more through welfare programs, crime, etc. than they add in economic benefits.
Posted by: Mark at May 23, 2006 1:56:19 AM
American farmers and farmers in CA are 2 different things.
Especially when I read that some of the vinyard owners were complaining if they
didn't have labor, they'd have to mechanize like others in the world. Their
grapes are too delicate to have to mechanize picking.
Seems to me more cachet than necessity.
Posted by: Sandy P at May 23, 2006 2:12:29 AM
Mark the non-economist Sailer seems to get the economics of this better than many trained economists:
1. Of course the “economical” definition of technology (“ideas or techniques to combine labor and capital in more efficient ways”) is also likely be be effected by the price of labour.
Are you seriously claiming firms do not take into consideration the demand for their products when they make investment in developing new ideas? Are entrepreneurs more or less likely to make labour saving devices if the price of labour is cheap?
Now this in itself does not mean growth is slowed by cheap labour, as I wrote in another post it is a question of how important you think externalities are (the more important the worse is illegal immigration) and how elastic the supply of research is (the more elastic the wrose illegal immigration.
2. What Sailer is saying is the question of SUBSTITUTABILITY. When you try to compare the costs and benefits of cheap labour in the US you are looking at a world where labour is cheap, and where labour saving technology has not been developed (or implemented, if otherwise available).
If the substitutability is high, as Sailer proposes it is, cheap labour is only marginally beneficial. The “country will stop without Latinos” argument works because many firms around, people tend to assume that’s the only way to do things.
In perfect markets this is not an important issue, the market will choice the cheapest way to do things.
But what if there are also costs associated with unskilled labour, but not it’s substitute capital?
Robotic car-washes may be slightly more expensive, but 21% of them do not receive Medicaid, they do not commit crime at 4 times the white-non Hispanic average, they do not vote for populists and leftist to raise taxes.
Now in the marketplace when firms determine what technology to implement they do not take these other costs into consideration. The carwash and the costumers benefits from cheap labour, while the taxpayers and community pay the cost. This is a classical problem in economics when you have incomplete markets, which is certainly the case regarding crime and the political system.
If the supply elasticity of capital is high and capital is a complement to unskilled labour the calculations of Borjas somewhat overestimate the negative impact of immigration on American workers. But if capital is elastic and a strong Substitute for unskilled labour we are likely to instead overestimate the BENEFITS of unskilled immigration!
This is in the end an empirical question, and one whose answer I have not seen.
But to the pro-immigration crowd: doesn’t it at least hurt your professional pride as economist that you need a journalist and Conservative movie critique to point this obvious fact out for you, albeit in non technical common sense terms?
Posted by: Teller at May 23, 2006 3:08:13 AM
This is what the greatest public intellectual of the previous century had to say about the subject:
Milton Friedman, 1999
"Q: Dr. Friedman should the U.S.A. open its borders to all immigrants? What is your opinion on that?
A: Unfortunately no. You cannot simultaneously have free immigration
and a welfare state.
Q: Do you oppose a unilateral reduction of tariffs and if not how can you oppose open immigration until the welfare state is eliminated?
A: I am in favor of the unilateral reduction of tariffs, but the movement of goods is a substitute for the movement of people. *As long as you have a welfare state, I do not believe you can have a unilateral open immigration.* I would like to see a world in which you could have open immigration, but stop kidding yourselves. On the other hand, the welfare state does not prevent unilateral free trade. I believe that they are in different categories."
All pro-immigration economist have responsibillity to answer this question: are immigrants like goods? Can we make an economic analysis with this narrow focus? How can open borders be combined with a welfare state?
Posted by: Teller at May 23, 2006 3:26:54 AM
Mechanizing vine cultivation -- not merely harvest.
Posted by: Mitchell Young at May 23, 2006 11:01:05 AM
Steve writes: "One thing that's clear is that illegal aliens are a substitute for investing in labor-saving technology, just as an ample supply of slaves was a major reason for the lack of technological progress in the Roman Empire."
This logic was taken to the extreme by the National Party in South Africa, after it had all but abandoned Apartheid by the late 70's and surreptitiously started opting for a policy of racial integration: they actually forbade motor car manufacturers the importation of automated machinery, hoping to prevent job losses and more social unrest. Of course, it only made the industry uncompetitive and cars hideously expensive.
Having been convinced by the mining lobby that the economy could not succeed without black labour, the Nationalists in the end had to shore up the economy in order to help blacks succeed. How long before America catches up with that beacon of progress, South Africa?
Posted by: Will at May 23, 2006 12:43:29 PM
"Adoption of labor-saving devices (capital accumulation) will be affected by immigration but that's not what drives economic growth. The fact that immigrants wash your car instead of some $100,000+ machine does not, alone, affect the living standards of Americans."
It's been awhile since I studied growth modeling, but this sounds about the opposite of the truth. Increasing population (via immigration) reduces the amount of capital per worker and therefore lowers the amount of income per worker (which is a reasonable measure of the "quality of life"). See the relation between income and population growth across countries.
Further, the amount of capital per worker is NOT irrelevant to living standards; though long term economic GROWTH is, as you say, determined by increases in technology once the economy has reached its "steady state." That is, at the steady state depreciation outpaces increases in the capital stock, bringing the level of capital back down to its equillibrium. But this "steady state" is determined by the savings rate, and the savings rate can be effected by investment demand. Clearly limiting immigration would increase the demand for certain kinds of capital, which would cause the savings rate to increase; thus shifting the "steady state" to a different level, a level with a higher capital to labor ratio and a higher standard of living.
Further, you seem to neglect the fact that fewer immigrants would increase the amount of technological innovation--"necessity is the mother of invention." Today there isn't a necessity because we have plenty of poor, uneducated immigrants working for peanuts. This could be holding back innovation, acting as a sort of developmental throwback. Also, I don't see capital accumulation and technological progress being mutually exclusive; the more reliant we are on capital the more incentive there is to improve that capital. Like I said, it's been a long while since I studied economic growth, so if I am in great error, please feel free to critique.
Posted by: Hans Gruber at May 23, 2006 1:39:45 PM
Don't forget, Juarez is (was?) a major manufacturing center with its maquila industry. Not to mention, well, a lot of us were already in the area before the gringos showed up (nothing like the annual friendly debate with Plymouth over the REAL First Thankgiving, held by de Onate several decades before there English speakers on this continent), and we do like to congregate where the good fiestas are gonna be.
Posted by: agm at May 23, 2006 7:24:00 PM
Also it might be useful to recall that El Paso has a university fairly well respected south of the border (UTEP was rather proud of the fact that at one time most of the practitioners of a certain branch of engineering in Mexico were trained in the one place; the College of Engineering is still pretty good), as well as proximity to one of the major settlements in the territories once known as Northern Mexico. Juarez (at the time El Paso del Norte) was settled where it was because of the awe-inspiring agricultural fertility of the region at the time, especially in comparison to the desert one had to trek northwards across to reach the northern territories.
Accidents of history and geography are probably important in explaining why El Paso draws so many despite the fairly stagnant local economy.
Posted by: agm at May 23, 2006 7:31:08 PM
One thing I really can't believe is this Mexican Tourism ad from Jones American Comedy Network. "Experience Mexico the way they experience the USA". They have it in their podcast too.
http://americancomedynetwork.com/FLASH/MexTourism.htm
Posted by: Steve Waikley at May 23, 2006 8:27:00 PM
Teller
Half the population of the US is already Southern, Black, or Hispanic. And about
three quarters of the children. Deal with it.
Posted by: wkwillis at May 24, 2006 6:31:46 AM
http://www.ailf.org/ipc/policy_reports_2003_SuccessOrStagnation.asp
was an earlier paper on Latino assimilation over time. The paper does not control
for the systematic expulsion of poor latinos from the US during the depression.
Deporting the poorest immigrants will indeed make it look as though the immigrants are assimilating be skewing the income distribution.
Posted by: wkwillis at May 24, 2006 6:38:24 AM
Sorry for veering off-topic, but I would like to see the immigrants-don't-affect-wages crowd discuss today's NYT article about lifting immigration limits for nurses. http://www.nytimes.com/2006/05/24/world/americas/24nurses.html?pagewanted=2&_r=1 How could this not affect wages of American nurses?
Posted by: FE at May 24, 2006 7:32:03 PM
So as a result of unusually high immigration rates in recent decades, the minorities will be half the people in the future,..... and already over half of the American population is Black, Hispanic, or poor southerners.
Can you say "Third World Country"?
Posted by: Brent at Sep 5, 2006 3:08:28 AM
interraciaal sex ^^^ jong actie in bad ^^^ bionde amore in casa ^^^ loghi suonerie gallerie ^^^ latex gay ^^^ chauve nympho leche le cul ^^^ seins quarante ^^^ quarante et un menue ^^^ pallido allievo figa fotti ^^^ deciso bionde inculate ^^^ modig skolflicka onanera ^^^ ohygglig amator karlek ^^^ damerino cowgirl prostituta ^^^ carino fighette sex ^^^ concombre chatte omasex ^^^ devergondee nue porn ^^^ lyijykyna ylaosaton porno ^^^ asiants imu penis ^^^ varm heldig jomfru ^^^ kaldere strand ^^^ modig klipp ^^^ kaldere lesbisk ^^^ haritomeni astinomikos ^^^ mathitria kolos stin kouzina ^^^ fils petillant ^^^ plus frais pere photos ^^^ to pio kafto tolmiros kori ^^^ pio kafto tolmiros kori ^^^
Posted by: levan at Sep 8, 2006 4:45:46 AM
