Wednesday, March 22, 2006
National Press Club
The report we're releasing today challenges this idea that there is such a thing as work that Americans won't do. And I'm not going to summarize it for you Steve Camarota, the author, will go into it but I think this is one of the important issues that simply isn't dealt with. It's just taken for granted that it's true when it's not. And as Will Rogers said, it's not so much what people don't know that's a problem; it's what they do know that just isn't true that's the real source of problems. So we're releasing today a report on this subject. You all have copies from outside. Take more if you want them. And it's online in its entirety already at CIS.org.
The speakers will be first the author and then two respondents. Steve Camarota, the author, is director of research at the Center for Immigration Studies one of the top people in the country on the economic and demographic and other impacts of immigration on the United States. The two respondents will be Jared Bernstein from the Economic Policy Institute, where he is director of the Living Standards Program he has been there since 1992 former economist at the Labor Department and author of a variety of publications, co-author of The State of Working America, an annual publication that comes out from his organization; and the other respondent we're happy to have from Boston is Paul Harrington, associate director of the Center for Labor Market Studies at Northeastern University. He's taught economics for 25 years, and his center examines many of these kinds of similar questions: the youth unemployment, the immigrant labor force in New England, et cetera.
So what we're going to do is Steve is going to give a presentation summarizing some of the findings in his report, we'll have responses from the two commentators, and then we'll take your questions and have some discussion.
As Mark indicated, one of the central aspects of the ongoing debate over immigration is the economy and the labor market. It's at the very heart of what's being discussed right now today, in fact in the Senate. The Senate in general, whether [senators such as] McCain or Kennedy or Arlen Specter, wants to legalize illegal aliens here, increase legal permanent immigration, and create a large new guestworker program. The senators have generally decided, or accepted, the view of the business community that America is desperately short of less-educated workers to fill low-wage jobs that require relatively little education. This of course is very germane in illegal immigration in particular because most estimates agree that about 80 percent of illegal aliens have no more than a high school degree in terms of their education. That is, they have no education beyond high school 80 percent.
The question is and what this report that we're releasing today tries to answer is, is this view correct? Are we really desperately short of what we might call less-educated workers? Now, to answer this question we used what's called the Current Population Survey, which is gathered on a monthly basis by the Census Bureau. We used the March file because it over-samples minorities and is usually considered the best file for looking at the foreign-born. And, very briefly, we define the foreign-born as people who were not U.S. citizens at birth. So these are people born in other countries that have come here. People born in, say, Puerto Rico are considered natives or native-born, and so are people born abroad of American parents.
Now, we focus our analysis on people who are 18 to 64 years of age. This is the most important group. They make up the vast majority of workers. They make up the majority of productive workers. They provide most of the household income in American families. So how they do in the labor market, the 18-to-64-year-olds, is obviously one of the most important things.
Now, we examined various ways in which they're attached to the labor market how they're doing. We looked at the share who are holding a job. We also looked at the share who are unemployed. And we also looked at the share that has left the labor market, or is not in the labor market. People not in the labor market are neither working nor are they looking for work, and we focus, again, on the 18-to-64-year-olds.
Well, I would say that we have basically four key or four main findings, and I'll run through them very briefly. Based on the data that the government collects, we find that there is a huge number of natives with relatively little education in the United States. There are 65 million native-born Americans who have no more than a high school education in the United States in 2005. Nearly 4 million of these 18-to-64-year-olds with no more than a high school education are unemployed. In addition, there are 19 million additional workers additional individuals 18 to 64 who are not even in the labor market. That is, they're not working nor are they even looking for work, so the idea that there is just nobody there doesn't really make any sense. There does seem to be a huge number of natives with relatively little education and again, that really isn't in dispute.
Now, the second key finding is also not really in dispute either: The share of less-educated workers has not done well in the labor market in the last five years. Unemployment among workers with no more than a high school degree has increased by 1 million between 2000 and 2005. The share not in the labor force at all again, not working, not looking for work is up 1.5 million. But perhaps of greatest concern, the percentages of adult natives 18 to 64 who are in the labor force have been declining.
Looking first at those who haven't completed high school the high school dropouts we find that the share who held a job between 2000 and 2005 declined from 59 percent to 56 percent. And of natives who are adults 18 to 64 who have a high school degree but no additional schooling, the share holding a job declined from 78 percent to 75 percent. And this trend continued through this whole five-year period. There has not been an upturn. In fact, one of the most troubling things is that even through January 2006, the share of dropouts and the share of natives with only a high school education who are not in the workforce or, that is, who are in the workforce hasn't recovered at all despite a general upturn in the economy. Something is going on. Again, this is not in dispute.
Natives with relatively little education their unemployment is up and their workforce participation is down. And it has not improved, for the most part, even through January of 2006. Now, I think these basic two points should not be we shouldn't want to debate them. There are lots of unskilled natives or less-educated natives and they've not done well in the last five years.
Now, the third finding I think is also important, and I think most economists agree with this point: The idea that there are jobs that natives don't do is simply wrong. Or maybe put a different way, the idea that the economy is divided between occupations and job categories that only immigrants do and jobs that only natives do is simply not correct. If we look at Table 5 in the report, for example, we look at the five occupational categories in the United States that have the most immigrants. Well, what we find is these occupations that have the most immigrants still have 22 million native-born Americans in occupations with very large immigrant shares. If we just focus on less-educated natives that is, those with no more than a high school degree we can see in Table 6 that there were 14 or more than 14 million in these heavy immigrant occupations.
Now, if the idea was that natives and immigrants did completely different kinds of jobs and they never really competed, we would expect to find virtually no natives in those occupational categories, but in fact we find 14 million less-educated natives in occupations with the most immigrants, and we also find another 1.7 million natives who are unemployed in those same occupations. Now, we also, in Table 10, look at illegals in these top occupations as well, but the main point to take there is, again, there are still millions of less-educated natives in these same occupations who seem to be in direct competition with less-educated illegal immigrants.
Now, you can think about the economy as divided into more broad occupational categories like construction, or you can try to look at great detail at very specific occupations, and one of the things we do in this report is we try to look at all 473 occupations used by the Department of Labor and Department of Commerce the way they divide the total economy and look at what share of each of those occupations is foreign-born, is immigrant. Now, the data that we're looking at there would be both legal and illegal. I should make that clear. Illegals do show up in most of the government surveys. Most researchers, including myself, people at the Pew Hispanic Center, Urban Institute, think that about 90 percent of the illegals do show up in government surveys of this kind.
When we do highly detailed analysis and you can see that in the last table of the report, Table D. It turns out that there is virtually no majority immigrant occupations in America. Again, if immigrants do jobs that natives don't, if the economy really is divided in that way, we should find occupations that are 80, 90 percent immigrant with virtually no natives in them. Those jobs don't exist.
Some occupations do have a large share of immigrants, but still, there are only 35 of the 473 detailed occupations in which one-third of the workers are immigrants, and these occupations account for less than 7 percent of the U.S. workforce, and I should add that there are over 5 million natives employed in even these most heavy detailed immigrant occupations.
Our analysis shows that when you look at the most detailed occupations, about one-half of native-born dropouts and about one-third of natives with only a high school degree and no additional schooling are in occupations that are about 15 percent immigrant. So that does mean that not every unskilled natives faces a lot of job competition from immigrants, but what it does mean is that a very large share of adult natives 18 to 64 who have relatively little education do face very significant job competition or, put simply, most of the construction laborers, cab drivers, nannies, maids, housekeepers in America are native.
A lot of times you still, though, even despite this data, will have people say things like, well, immigrants, really they only do jobs that natives don't want, but most of the people who say that are more educated Americans and more affluent Americans. And what they really mean when they say immigrants only do jobs that natives don't want is that they do jobs that I don't want as a more educated and affluent American.
Now, the fourth finding is one of the most difficult to get our handles on. We try to look at the evidence for whether immigration explains this decline. As I said, no one disputes that less-educated natives have done very poorly in the last five years. Again, the debate in the Senate, you would never know that from listening to it on immigration. The possible harm done to natives never comes up in the Senate. It did come up in the House debate. Oddly enough it was the Republicans who brought it up mostly last year, but this year in the Senate the Democrats and Republicans never talk about less-educated natives and the harm immigration may do to them, even though the evidence is overwhelming that that group of workers has done very poorly.
Now, what did we find? Well, we did find that states that saw the biggest growth in their immigrant workforce generally did see the biggest declines in less-educated employment. Just to give you an example, in states that had statistically significant increases in the number of immigrant workers, the share of natives who are less educated holding a job declined by five percentage points. That compares to the overall decline nationally of about 3 percentage points. So what that tells us is that we do have some evidence that less-educated natives tended to leave the labor market more in high-immigrant states.
We also find that in occupations when we look at occupational categories that occupations that have the most recent immigrants, in general they're also occupations where native unemployment tends to be the highest. We also looked at it by age group, divided it by education. Again, in those age groups in general where you saw the biggest growth in immigrants as a share, you saw the biggest declines among natives not in every case. But in general, when we look at those sectors of the economy or those parts of the country that had the biggest influx of immigrants, we generally find that natives did the worst in those areas.
Now, why might that be? Why would immigration affect natives? Well, as I've said, they seem to do exactly the same kind of work. Less-educated natives and less-educated immigrants do often the same work, so they will be competing for jobs, so that's an important point to make. It could be because the immigrants come in and might be willing to work for less. That possibility exists there is some evidence for that so that they crowd out the natives, they get a leg up on the natives. Employers see that; they like that. But the evidence that immigrants work for less is not that strong. Immigrants make a lot less than natives, to be sure. But they mainly make a lot less because they're a lot less educated on average. Unskilled immigrants or immigrants with relatively little education who are recent arrivals don't make much money, actually regardless of legal status, but it's their educational attainment and set of job skills that are the main impediment in the labor market, not that they're foreign-born, not their illegal status.
There has been work on illegals, if you're interested. In the 1980s we legalized about 3 million illegal aliens, and we watched very carefully what happened to their wages immediately thereafter, and their wages went up maybe 5 percent. So that sort of research suggests that the penalty for being illegal in the United States is maybe 5 percent not zero but hardly a huge difference.
But there is another possibility I think is more likely, is that immigration does significantly increasing the supply of workers, the number of workers does probably lower wages and benefits and working conditions, and that may also make a job less attractive to natives. That can sometimes happen. But it's not because they come in and work for less; it's just their mere presence in the United States. Increase the supply of workers; reduce wages.
There is also a very real possibility and a lot of anecdotal evidence that immigrants are perceived of as better workers. Especially if you turn off the microphone, big and small employers will tell you they think this. And unfortunately, they also think this especially when it comes to, say, African American workers. The perception is that immigrant workers are better than black workers. There has been sociological and anthropological research on it, but it's not that hard to find employers who unfortunately hold such views.
Another reason why immigrants might crowd out natives is that they may possess more efficient networks of hiring and employment. In the old days, an employer may have put the ad in the newspaper and put up a poster at the local supermarket indicating that, hey, he's got another job or two, but once maybe he hires a foreign-born foreman, that foreman kind of takes over that function. He just tells everyone the employer tells him and then that foreign-born foreman just tells everyone in his apartment building. The foreman just tells his brother-in-law and his cousin about the new jobs. And in that situation, the natives kind of get shut out because they're still looking in the newspaper, which the immigrants are too, and they're looking in the Penny Saver and the posters in the supermarket, but in that situation you would have the immigrants would have a leg up through a kind of well developed network.
Well, let me conclude here by restating some important points. In 2005 there were nearly 4 million unemployed native-born adults in the United States who had no more than a high school degree. There was another 19 million who are not in the labor market. That means they're not working and they're not even looking for work. Perhaps most important, the share and number of people who are unemployed who are less educated in the United States is way up. In fact, the whole idea and this comes up in the debate that America has a 5 percent unemployment rate and therefore clearly we need lots of immigrant labor is very foolish. It's irrelevant to the debate over illegal immigration and unskilled immigration in general because unemployment is not 5 percent at the bottom end of the labor market. In high-immigrant occupations it's 11 percent. Among natives with less than a high school degree, it's 13 percent. That is the more relevant number when we're talking about, is America desperately short of unskilled workers or less-educated workers?
As I said, the labor force participation of less-educated workers has fallen significantly, and the decline is particularly troubling because these workers already have the lowest workforce participation rate, they already have the highest unemployment rate, and they already have the lowest wage rates. So the last five years you took the poorest workers and made them even worse off.
The available evidence does show that less-educated natives and less-educated immigrants, legal and illegal, do exactly the same kinds of jobs. There are no such things as jobs natives don't do, even when we look at all 473 detailed occupations used by the Department of Commerce and Labor. We also find evidence that in those areas of the country and in those sectors of the economy with the biggest growth in immigrants also saw the biggest declines in native employment in terms of those holding a job, in terms of those unemployed, and in terms of those who are part of the workforce.
In the debate in the Senate going on right now, the impact of immigration on America's most vulnerable workers unfortunately has not come up very much, yet it should. Letting illegal aliens stay in the country, dramatically increasing the number of green cards, as Arlen Specter and Senator McCain and Senator Kennedy want to do, and creating a new guest worker program, as they want to do, almost certainly has enormous implications for America's poorest workers. This research suggests that the impact on those workers should be a central part of the immigration debate.
Now, one could still favor legalizing illegal aliens or some increase in legal immigration for some other reason, but it is extraordinarily hard to make the case that America is desperately short of less-educated workers, given the enormous numbers and the deterioration I've been talking about. The government's own data make this fact abundantly clear.
MR. KRIKORIAN: Thanks, Steve. Let's start with Jared Bernstein.
JARED BERNSTEIN: Thank you. Thank you for inviting me. Thanks to Steve for inviting me to be on the panel. He is truly an expert in this area of work, and while I don't agree with everything he says, I've learned a lot from him about studying this issue.
I have a handout that I passed out. Let me make sure my fellow panel members have it. I believe you all have it; if not, someone nearby you should have it, and I will refer to the figures in the handout.
Are foreign-born workers displacing natives, or do the experiences of most workers in recent years, regardless of their nativity, suggest an underlying weakness in labor demand and insufficient job creation is the root problem facing workers in this country, immigrants or natives? Now, another way of putting it is, for native workers is the problem immigrant supply, immigrant competition, or is it weak labor demand, and I believe it's largely the latter. That doesn't mean there is no economic competition between immigrants and natives; nor does it imply that supply effects, which Steve talked about, the increased supply of labor putting downward pressure on wages and living standards should not be taken to imply that such a thing never exists. But over the period of Steve's study, which is pretty much 2000 to 2005, I think the problem is more insufficient job creation than it is immigrant competition.
Turn to my Figure 1, please. And this is using a slightly different dataset than Steve uses, but the results are almost exactly the same, so that part doesn't matter. It's just that some of the numbers that you see in my charts might differ a little from his but only a little, and in no case are the results significantly different.
What you see here in Figure 1 is the change in employment rates for all the groups that Steve looks at in his study. I also looked at 18-to-64-year-olds. It's the same sample. In fact, it's a similar underlying dataset, and you see all workers, all foreign-born and all natives, and then we follow that pattern for each educational category. And if you just kind of hold this thing up and take a look at it, it basically is a picture of a lousy job market. Between 2000 and 2005, the employment rates you know, that's the share of the population that's employed for each one of these groups was either flat or falling. And so this is a picture of a uniquely tough job market for everyone, regardless of nativity. And what's kind of remarkable is that four years of this is an economic recovery that's been fairly robust by some measures. And of course embedded in this is the jobless recovery, the longest one on record.
Now, as Steve will correctly point out, and emphasizes in his piece, in every case native employment rates fall more than those of immigrants, and so there is a difference there that we need to explore. Is this prima facie evidence of competition and displacement? Well, not so if they�re largely in different industries, occupations, or different parts of the country, both of which are to some extent the case but not as Steve points out not wholly the case by any means. And by the way, this is one reason why some studies find only negligible effects on wages of immigrant competition. Other studies find larger effects.
If you look at the Congressional Budget Office's recent piece on immigration, you will find that and in fact, it's quoted in a Bob Samuelson piece today in The Post, that takes off on some of Steve's work, and quotes the CBO study saying, Some studies find negligible effects, some studies find effects that are much larger, in terms of the wage-dampening effect that immigration supplied. And one reason that the studies on the negligible end of the continuum will find that is because in some parts of the country, in some occupations, immigrants seem to be more competing with each other than with natives. In fact, one of the few sectors with consistently strong demand over the period that Steve is focusing on has been construction, of course, an industry with high immigrant penetration, particularly in some parts of the country. And if you look at Figure 2, you'll see the point that I'm trying to make here.
Here I just indexed the employment growth in construction and in non-durable manufacturing, and in non-durable manufacturing I tried to take out the sectors that are heavily immigrant food, apparel, textiles. And the idea here is that an industry heavily populated by immigrants did uniquely well over this period. And I chose non-durable manufacturing because we're really trying to look at the experience of low-skilled workers here, and while non-durable manufacturing fell by 10 percent, construction is up by 10 percent in terms of employment. So I think one of the things you're picking up here is really demand-side changes having to do with insufficient job creation, particularly in industries and occupations more dominated by natives.
Two other points regarding this difference in employment rates. First, you have to recognize that the results in Steve's report are generally net results. That is, you take the growth flows of job creation and job destruction and you subtract them out; you get the net flows. One needs to be careful in this kind of research not to suggest that no native workers got jobs over this period. I mean, think about where you work. Were there any non-immigrants hired there over the past five years? The net result is of course important and it's indicative of weak job creation, but certainly some natives got jobs over this period. I mean, just looking at the college employment record, 3 million more natives were employed college graduates 3 million more native college graduates were employed in '05 than in '00; 1 million more natives were employed in '05 than in '00, but the employment rates went the way they did because that employment growth record was not sufficient to boost employment-to-population ratios.
Okay, so actually I'll just move on a little more quickly here. Secondly, the underlying numbers are much larger for natives, of course, than immigrants, so in numerical terms there's no way you can explain a one-for-one tradeoff between native and immigrant differences in employment rates. In other words, even if you believe that immigrant competition is an explanation for what you see in that first figure, it's got to be only partial because the difference in the employment rates is too large to be explained simply by the immigrants. There are just not enough in those categories to close those gaps.
Okay, so demand matters, too, and immigrant I talked about the immigrant industries. All that said, Steve and others are correct to raise the point that if there is less competition between immigrants and natives in strong job markets, shouldn't there be more in weak job markets? Now, let me give you a little background for that point. Over the course of the 1990s, particularly the latter 1990s, we had a really remarkable period although it really shouldn't be that remarkable it was really the first time we had a full employment labor market in 30 years. And this was a period as I�ve learned from some of Steve's numbers, this was a period of large historically large immigrant inflows over the course of the 1990s. I think it's the decade with the largest numerical increase in immigration.
And over that decade, both natives and immigrants alike were amply absorbed into the labor market and saw their wages, incomes, living standards rise, their poverty rate shrink at historically fast rates. So that period serves as, I think, an important benchmark for a labor market that;s creating sufficient demand to absorb both immigrant and native supplies. So that said just getting back to this question I raise that's a strong labor market. Given a weaker labor market, aren't you going to see more competition? Don't you have a larger labor supply competing for fewer jobs? If you hold a press conference like this in a small room, you're going to have people standing up. So part of this is just a weak you're looking at a weaker job market with a consistently growing supply of labor.
Well, I gave you the caveats about sectoral concentration, which dampens that kind of competition, and it's likely the case but it's likely the case, especially for the least skilled, that maybe behind the fairly dramatic difference in the employment outcomes for high school dropouts and if you look at Steve's numbers, here is where you see, I think, the largest difference between the experience of natives and immigrants. Native high school dropouts in my data was about the same as his. Native high school the employment rates of native high school dropouts rise a little bit. The employment rate I'm sorry, the employment rates of immigrant high school dropouts rise a little bit; the employment rates of native high school dropouts fall about 5 percentage points, which is large for a five-year period.
Now, here I have another set of caveats Figure 3. Figure 3 shows the share of high school dropouts as a share of the population, and that first bar is simply comprised of the the components of the first bar are the second two bars. So, in other words, there is 14.1 percent of the population are high school dropouts in 2000 those are the first three bars 4.5 of that is immigrants, 9.6 are natives, okay? So it's 4.5 plus 9.6 equals 14.1. That's your high school dropouts as a share of the population. And there are a couple of important points here. First of all, it's a small sector. It's not miniscule, it's not unimportant, it's not tiny, but it's a small sector as a share of the labor force 14 to 13 points and it's a shrinking sector.
What's important here is that the immigrant share of that sector is growing; the native share is dropping. The native share is shrinking while the immigrant share is rising. Over this period there are 1 million more immigrant high school dropouts and about 800,000 fewer native dropouts. My point here is that if you're a native-born person without a high school degree, you've got a whole slew of problems and immigrant competition is but one of them. Would your economic fate be better in the absence of competition from similarly unskilled immigrants? Probably for some it would, but take a look at Figure 4. This comes right out of this, I thought, very piercing article in the New York Times this week about the plight of male high school dropouts by race. I mean, just look at the first bar. This comes right out of that piece. I mean, it says that 72 percent of young male high school dropouts in 2004 were either unemployed, not in the labor force, or in jail.
I mean, you know the statistic that on a given day there are more young high school dropout black men in jail than there are employed. If you're a member of this group, you've got a whole set of barriers and immigrant competition is merely one of them. Does anyone think that were immigrant high school dropouts to disappear from the scene tomorrow, this group would be much less disadvantaged?
Okay, I have a few more minutes, and I'm coming to the end so I think the timing is going to work out here.
I performed a simple statistical analysis comparing the changes in the employment rates among native workers with the change in state immigrant shares. And this is in Figure 5, and I just want to kind of close with this. It's sort of a punch line. Let me just go over that one more time. Steve actually talked about a similar exercise. If you take each state and you look at the change from 2000 to 05 in the employment rates of native workers and you compare that using regression analysis to the change in immigrants shares, here is what you find. In the interest of time forget the first bar; just look at the second two bars. The change in immigrant share gives you a coefficient that's negative and significant, and it suggests that a one percentage point growth in the immigrant share in a state leads to about a .25 percentage point decline in the population share of natives.
A couple of things to recognize here. The coefficient is significant and negative, but it's far from a one-to-one tradeoff. Overall, if you believe this result and you've changed the rate overall if you believe this result and you don't have any growth in immigrant shares over this period, native employment rate loss, instead of being 2.5 percent, would have been 2.2 percent, so it wouldn't have changed the outcome very much. However, once you control for demand and I do that by just controlling for the manufacturing share of employment in the state the coefficient becomes insignificant and the relationship that Steve is emphasizing goes away. So again, I think you don't find this relationship once you take into account changes in demand over this period.
Okay, so concluding. What we have here is an underperforming labor market for all comers. What these results say to me is that not we suffer excess immigrant competition, but from a lack of a full employment labor market, the type that prevailed in the latter 90s. Now, you might have a case here for high school dropouts, but native high school dropouts face a slew of challenges of which immigrant competition is but one. I'm sure we'll talk about policy implications during the Q&A so I'll cut that part out here. I'll just say this for now: The bills under discussion have an important and useful purpose of trying to rationalize an irrational and dysfunctional system. I'm totally with Steve on this point that it's wrong-headed and non-economic to argue that there are jobs Americans won't do. I don't believe anyone's interests are served when we nod and wink at 11 million undocumented persons who employers are free to hire without fear of sanction. However, not all legislation is created equal. There is a continuum with amnesty on one end and hyper-punitive approaches on the other.
The likely result, I think and I hope, will be some type of compromise in the spirit of Kennedy-McCain and we can discuss the relative merits later. I'll just note here that from the perspective of labor market outcomes, bringing undocumented workers out of the shadows and under the umbrella of labor protections, including minimum wage, overtime and perhaps union coverage should help diminish their exploitation and raise the competitive playing field for all workers.
MR. KRIKORIAN: Thanks, Jared. Paul Harrington?
PAUL HARRINGTON: Thank you. First let me begin by distinguishing between two different ways that we count employment in the United States. One is every month we do a count of jobs. We go to firms and we ask them about payroll employment. And what payroll employment refers to you'll have to excuse me; I'm a little froggy what payroll employment refers to are jobs where there is an employer-employee relationship, the firm engaged in unemployment insurance, Social Security tax withholding, federal tax withholding and the like. So it's a regular wage and salary job.
So when you report construction payrolls, manufacturing payrolls, that refers to that payroll, wage and salary survey. The data in most of the discussion you've heard today comes from another data source called the Current Population Survey, which measures number of people that are employed in a month. So there are two different measures going on. Generally the way it works is the counts of jobs is a subset of the count of people who are employed because there are a number of people that don't hold wage and salary jobs but who go to work. They're self-employed people and other kinds of workers. So the results of these data work in two different ways.
When you heard Mr. Bernstein's discussion he talked about the poor recovery in the job market. When you go back and you look historically, how well did the wage and salary recovery occur in the U.S. economy over the last four years? The answer is, by historical standards it was extremely poor. Look at the last five economic recoveries. During the first 15 quarters of those recoveries, wage and salary employment grew by about 10.4 percent. During this recover, wage and salary grew by only 2.3 percent. So the question is, is how do we get our unemployment rates down so much in the absence of this wage and salary employment growth?
And part of the answer is when you take a look at the findings of the current population survey, what you see is is that a lot of the employment growth that we had in the country really occurred outside of the wage and salary system. If you look at the end of the recession up through the third quarter of last year, the payroll employment levels grew by about 3 million. The number of people who said they were employed in the household survey grew by 6.3 million. So far more people said they were employed in the household survey increased in the household survey relative to what the wage and salary survey said.
So what's going on here? How do we reconcile this? Well, there's lots of different there's really three different explanations. One part of the explanation is we had a rise in legitimate self-employment, people that were true entrepreneurs going out to work. Second is that we had a rise in kind of contract contingent workers you know, 1099 workers, people that they look like they have a wage and salary relationship except they're not on their firm's payrolls, and we're seeing, maybe in your business a little bit, some of that happening, certainly in the finance industry and the like.
But then the third thing that we're seeing is a very large increase in the number of recent immigrants who become employed in the U.S. One of the things that we do in the household survey is in addition to asking you where you were born, what country you were born in, we ask you when you came into the country, and then we also ask you about your labor force status. When you take a look at the data for people between who came into the United States between 2000 and 2005, what you see is something like this, that during that 2000 to 2005 period now remember, we went into recession and then we went into recovery between 2000 and 2005, the number of people the rise in employment, in employed persons in the United States, was about 3.96 million. The number of new immigrants, people that came in between 2000 and 2005 who got jobs during that period of time, was 4 million.
What that says is that total net increase in employment over the last five years has been concentrated among new immigrants. So a lot of the reason why employment off the books went up and employment on the books didn't go up is because we've had this growth in the new immigrant population very rapid growth in the new-immigrant employed population.
When you take a look at the characteristics or who those individuals are, three things really jump out at you. One, young immigrants immigrants in the labor market tend to be young. Sixty percent of immigrants in the job market are under the age of 35. So they're very young relative to the native-born population, particularly in my part of the country where I'm still a youngster up there, believe it or not. The second thing you see is that they're overwhelmingly male. For every one new immigrant female that's in the job market, there are two males. So they're overwhelmingly male. So they're young and they're male. And then the third thing, as Steve mentioned, is they have very low levels of educational attainment very disproportionately concentrated among high school dropouts and high school graduates.
Now, when you take a look at the nature of job growth we've had in the American economy over the last 20 to 25 years, you see that an overwhelming fraction of the net new job creation we've had has been in the college labor market. About two-thirds of all the employment rise we've had has been in the college labor market. There are some moms and dads in this room who have written some pretty big tuition checks, and the reason why they've done that is because the size in the earnings premium for college grads has risen so rapidly over the last 25 years. So today a young person graduating from college could expect to earn somewhere around 70 percent more per year than the young person graduating out of high school.
But when you take a look at the data that Steve put together on the occupational composition of the immigrants that are employed in the job market, very heavy concentrations among high school dropouts, very heavy concentrations in low-end occupational areas, so that there is a very substantial disconnect, based on the evidence that he has, that suggests that the growth sectors of the American economy are not at all being served by many of the new immigrants, particularly illegal immigrants that are coming in in these low-end labor market segments with very low levels of educational attainment.
The Bureau of Labor Statistics and I worked for them many, many years ago just recently released their projections of employment by occupation and by educational attainment, and when you take a look at the BLS forecast, what you see is that they expect that the rate of demand for growth among college graduates will be double the rate of growth for demand for people who have a high school diploma or less. They expect a 20-percent rise through 2014 in the demand for people with Bachelor's degrees, a 10-percent in demand for people with a high school diploma or less. So the job content of the economy seems to be moving in one direction and the educational structure of the new immigrant population seems to be moving in another direction.
Part of the evidence that we put together in my center is is we believe that a lot of substitution that Steve has talked about has really been occurring among the under-30 population, that when you look at the decline in these employment population ratios you've heard and all they are is just what an E-to-P ratio just says is that if you've got a group of people, say, 16 to 19, what fraction of those are working at a point in time? And what you see is over the last few years, almost all the decline in the employment rates in the country, these E-to-P ratios, have been concentrated among people under the age of 30, and particularly in that 16-to-24-year-old population. Well, who are those 16-24-year-olds? Well, they're young men and women and they have low levels of educational attainment. They're high school dropouts; they haven't finished school. So that when you see the substitution you know, you see it where you buy your doughnut, you see it where you buy a cup of coffee, you see it in the retail trade stores. This is where that substation in large measure is occurring.
You know, one of the things that at the national level we don't have a job vacancy survey that measures job openings by occupation, so we can't do this we can't go back and take a look at the relationship between the number of unemployed workers we have and the number of job openings. But in my own state of Massachusetts, we do have an unemployment-to-job-vacancy measure. We measure job vacancies by occupation. In the immigrant-dominant fields that Steve talked about, particularly in construction, landscaping, some of the production jobs, some of the landscaping jobs, food service jobs, we find unemployment-to-job-vacancy ratios that are 10-to1 in other words, 10 unemployed workers for every one job opening. So to argue that there is any kind of labor shortage under those conditions would be just wildly crazed.
When you take a look at this evidence about this, you just see these labor-market imbalances it seems to me what we do is we flood the bottom of the labor market with some of our immigration policies, or lack of enforcement of policies, but then at the other end, at the college labor market where I think there would be a stronger case for immigration, we really haven't done much. Why? Well, because for an immigrant to work in the college labor market, they have to come through a legitimate pathway. They've got to come through a visa program, the post-secondary system some kind of legitimate pathway because you can't work as an engineer or an accountant otherwise.
So as a result, we're very effective at constraining supply at the high end because we enforce those rules of the labor market, and they're primarily enforced by the firms themselves. At the bottom of the labor market, what's happened is quite the opposite. And part of what I think and part of the reason I think that it reconciles some of Jared's findings and Steve's findings is I think the flow of immigrants themselves have begun to really erode fundamental labor market institutions, that the rules and regulations, the laws, the standards we've developed since really beginning with the New Deal have really begun to be fundamentally eroded as we flood the bottom, that no regulation can withstand very massive flows of immigrants and labor supply will always trump regulatory enforcement.
And the proof of this, I think, is those of you that live around here, take a drive out to the Little River Turnpike in Annandale. Go to Little River Turnpike and Heritage Drive, and at the corner there you'll see a 7-11, and there will be 70, 80 men, who I think probably come from Central America, standing around, and they'll be there early in the morning and they'll be there late at night. And they're waiting to get picked up, and they're going to get picked up and they're going to do construction jobs, landscaping jobs. Maybe somebody in a Mercedes will pick them up and have them put in a patio on the house; a construction worker will pick them up who knows? But all of this is employment that's off the books, unregulated. Young Americans cannot get access to these kind of employment opportunities.
And what is does is it fundamentally changes the nature of work, and that's part of the reason why I think we're not getting wage and salary growth, because we're pushing more and more employment over the last four or five years off to unregulated, undisciplined cowboy labor markets. And this is something I think in the long pull could be very destructive to the structure of opportunity in the United States.
MR. KRIKORIAN: Okay, thanks, Paul. I'll let Steve get a couple of responses in to the comments and then we'll go to Q&A. Steve?
MR. CAMAROTA: Well, I thank both panelists for their comments. Let me say this let me restate the numbers so we understand that it's true that there are fewer high school dropouts 18 to 64 who are native than there were five years ago, but their rate of employment and their unemployment rate are way up. Now, they faced problems in 2000 but things actually got worse, even though they were pretty bad in 2000.
MR. BERNSTEIN: You mean their employment rate is way down.
MR. CAMAROTA: Yes.
MR. BERNSTEIN: Okay.
MR. CAMAROTA: Their unemployment rate is up and their employment rate is down. There are, if you will, 450,000 missing high school dropouts 18 to 64 from the labor market. That is, had their labor force participation remained the same, there would have been 450,000 additional native-born high school dropouts in the labor market. There are a quarter of a million more unemployed. At the same time, immigration increased the number of foreign-born high school dropouts by 800,000. In other words, there are about 800,000 more native dropouts unemployed or not in the labor market, and there are about 800,000 more immigrants who are high school dropouts in the labor market. So we need to keep that in mind.
The other thing to point out is there are about 1.4 million natives who have only a high school degree who are missing from the labor market. Had their labor force participation rates remained the same, 1.4 million of them would have been should have been in the labor market. And there are about a growth of about 700 (thousand) or over 700,000 immigrants who have only a high school degree.
So Jared is right; I think demand is weak, but that strike me as a powerful argument to have less immigration. If you're not creating a lot of new jobs, it doesn't make sense to increase the number of high school dropouts by 800,000, especially when all the evidence shows native high school dropouts are doing so poorly. If natives with only high school degrees are doing poorly because there's not a heavy demand for their kind of work, why grow the number of immigrants with only a high school degree? It just doesn't make sense. It seems to me that if he's right and demand is weak, then that would strike me as a reason to go with an enforcement policy, and then if you want maybe the economy would pick up you could reevaluate your policy, but right now it just doesn't make any sense.
And I should point out that 2000 to 2005 appears to be the highest level of immigration, legal and illegal, in American history. According the current population survey, about 8 million legal and illegal immigrants settled in the United States each year. About 4 million were illegal and about 4 million were legal, according to my research and everyone else.
MR. KRIKORIAN: Over the whole period.
MR. CAMAROTA: I'm sorry, over the whole period 8 million, 2000 to 2005, highest in history, half illegal, half legal.
Now, Jared also does some regressions where he compares differences in employment rates between states. He finds some effects but then he's not sure. The reason I don't do that is that there is a problem, that a) states are not discreet labor markets. You can't say that all the workers in one state are in one labor market and all the workers in a different state are in another labor market. There is movement between states both in terms of people and goods and capital, so that's one problem. When the National Academy of Sciences looked at the question of immigration, it basically concluded that you couldn't compare differences too much between states or between labor markets because of the integrated nature of the national American economy.
The second problem with that is that in the current population survey every state has a margin of error around it, so you're running a regression that treats each state as what we call in economics a point estimate, but each state actually has a margin of error. So it's not clear that if you do or don't receive statistically significant results, whether you can make much inference from them. So that's why I avoid that in my analysis. But, you know, having said that, what I do is I just look at states that have statistically significant increases in the number of immigrant workers, and they do seem to have a significantly worse labor market outcome for natives than those that don't have statistically significant changes.
But in any event so I guess that's my response. And I would agree and let me make this point: I do not think that there is a one-for-one trade between jobs gained by immigrants represent a job loss by natives. That would be a mistake. But it would also be an equally powerful mistake to say that dramatically increasing the number of immigrant workers in the United States has no impact, or even a trivial or relatively small impact on less-educated workers who are native born. They do the same kinds of work, and their situation has deteriorated dramatically in the last five years during the last five years when we've had record legal and illegal immigration.
MR. KRIKORIAN: Okay, Steve. Let's take questions. Yes? Yeah, go ahead.
Q: I'm Lisa Sylvester with CNN. We often hear policymakers in Washington will say, well, we need to educate our people then. They often use that as we need to increase job skills. If this is the situation and native-born Americans are having a difficult time getting jobs and it happens to be the ones who are at the lower end of the education scale, then why not just educate the population?
MR. CAMAROTA: I would tend to be very supportive of job training. The results on that are somewhat mixed but I still think it's a very worthwhile thing. Educational opportunities are very important. But it seems to me that if you want adults who don't have much education to get more of it, the last thing in the world you want to do is push them out of the labor market. The last thing in the world you want to do is reduce their wages and benefits because how can they work part-time and then go back to school part-time if they're getting creamed in the labor market?
Making the low-income worker poor is not conducive to improving his situation, whether it's getting more training and skills or more experience in the labor market. One of the ways his situation would improve, hopefully, is he could get a job that would help retrain him a little bit, but if he's not getting a job because he's being crowded out in the labor market, that's not a very good deal. One of the ways he would get improvement would be he'd say, honey, you go back to school and I'll work more hours. But if his wages are super low, that's a very hard option to take for the spouse to go back to school. It seems to me that nothing could be more destructive to improving the plight of less-educated workers as to make them poor. It makes it harder for them not just to improve their lives but to improve their skills and education. I don't think that it's perfectly reasonable to say we should try and improve the skill set of our American workforce or natives or anyone in general, but it doesn't make sense to say that it's really helpful if we would only drive down their wages and reduce their job opportunities too; that would be really helpful to them. That doesn't really
MR. KRIKORIAN: I just wanted to add a quick point Jared had a comment too that your question isn't really just hypothetical. I mean, I've heard congressmen actually say, well, isn't it good to drive down the wages of Americans at the bottom of the labor market because it will give them an incentive to upskill to increase their education? I mean, this is actually a question this is the thinking behind some of the support for mass immigration. And in effect, I mean, the analogy I see is a guy who can't swim; you throw him in the water so he learns how to swim. Well, you hope that he learns how to swim before he drowns. But, I mean, it's disconnected from reality as far as I can tell.
MR. BERNSTEIN: Well, I think I'm probably closer to what I think was implicit in your question, which is the least-skilled worker the competitive burden for least-skilled workers is more their lack of skills than immigrant competition, and if you want to help that worker, there are a number of things you could do that we really haven't talked about on the panel today because they have nothing to do with regulating immigrant flows for example, raising the minimum wage, right? This is a policy that I think would push in the right direction on almost everything we've talked about thus far. I mean, if one of the things that inspires employers to hire low-wage workers is that the minimum wage is at a historical low, close to a 55-year low, raising the minimum wage would presumable go the other way.
Now, again, you have the illegal point. I mean, if illegal workers can skirt minimum wage laws, that hurts everyone. But I would think that if you were going to kind of look at the pantheon as I stressed in my talk, if you're going to look at the pantheon of barriers facing low-wage workers I mean, low-skilled workers, especially high school dropouts, the very greatest barrier they face is the fact that they're high-school dropouts.
MR. CAMAROTA: Paul, did you want to say anything?
MR. HARRINGTON: Yeah, I guess my only comment is I would disagree with Mr. Bernstein a little bit in that in a sense I think there is direct competition between foreign immigrants and particularly young adults. Again, a lot of substitution here I think is occurring among younger people, and so as a result and one of the things I find is early work experience economists find work experience itself has big economic payoffs, that independent of schooling, getting a kid a job just has some economic payoffs in terms of developing a whole set of non-cognitive kind of behavioral traits that employers value in the job market. And to the extent you get that substitution occurring in the job market, the productive capacity of those junctures really it will be diminished, you know.
MR. CAMAROTA: Let me talk about this question of the other things that impact less-educated natives. I think Jared is certainly correct that trade policy would be one, or that would be one, you might say. Or you might say that it seems also the case that the lack of skills, as he emphasized. Economists like to talk about skill bias technological change. It means as a technology changes, the value of less-educated workers just tends to go down because workers need to be able to use the new tools and technology in our economy, and that has nothing to do with immigration. Again, international trade and the internationalization of the economy, which tends to work against less-educated natives, also has an impact on less-educated natives, and it's probably significant.
But one of the distinctions about immigration is it is a choice. The United States cannot instruct the Japanese to stop setting up factories in China to compete with the United States. The United States cannot legislate a pause in the expansion of human knowledge, but it certainly can have less immigration. Polls show that's what most Americans would want. It's not the weather; it's not something outside of our control. Many of the things that harm less-educated workers in the United States are very difficult for us to do anything about. Immigration is not one of them. That is something we can do something about to improve the lives of low-skilled Americans if we have a view that that's something we want to do.
MR. KRIKORIAN: Yes, sir?
Q: Yeah, Tim Fry (sp) with (off mike) Broadcasting. Can you each of you who wants to respond to this tell us the impact of some of the the Senate compromise and some of the issues that are coming up with the guest worker program and that sort of thing on these people who have just a high school education or less, the native tell me what the impact of that is upon the native population.
MR. CAMAROTA: Well, I mean, the guest worker program that they have in mind seems to be very expansive hundreds and hundreds of thousand people or maybe even unlimited. It seems like the vast majority will have less than a high-school education. Of the illegal aliens here that they want to legalize, like Specter wants to legalize them and McCain wants to legalize them, and Kennedy, about 80 percent of those have no more than a high school education. And they want to also increase the number of green cards that's legal permanent residence.
Right now about half of new green card people who come in on green cards maybe have no more than a high school education. If they change the categories that probably would grow, so most of the people that would come in under the green cards because the countries that tend to probably would send more people. So the proposals of letting illegal aliens stay, dramatically increasing permanent immigration, and dramatically increasing guest worker immigration would significantly increase the number of less-educated workers in the United States with significant implications for less-educated natives. In other words, all the proposals envision big increases in less-skilled immigration or less-educated immigration, and that I think is the concern. And I wish that in the Senate they'd talk a little bit about this issue and at least recognize the plight of America's poorest workers.
MR. BERNSTEIN: Before I answer that, I just want to correct both Steve and Paul to some extent I think are mischaracterizing some of my statements. I've been very careful to take what I believe and hope is a nuanced view that at no point denies the possibility of immigrant competition with natives and in fact stresses that if that's happening anywhere it's among the least skilled. So I want to be clear. I mean, a couple of comments from Steve and Paul have suggested that I'm in this camp that is completely denying any of the possibilities they're talking about, and that's absolutely not the case.
The only point I would make to add to what Steve said two things. One is it's very possible that to the extent that workers who are now illegal and outside of some of the protections of labor laws that prevail in this country once they're brought in, that their wages and working conditions would improve. Certainly the unions are trying hard to organize these folks, and that would help to boost their earnings and living standards as well. The question is to what extent does a guest worker program signal people who are contemplating immigration that the doors are open; come on over? There is some suggestion that this is a positive elasticity there, that that would increase the flows. On the other hand, you know, the other side of that argument says the flows are already what they are and you're not going to change that, and the border is totally porous so such a signal wouldn't make any difference at all. And I don't know that anyone knows which of those is correct.
MR. KRIKORIAN: Could I just add one issue here is that the president's view of this I mean, he doesn't really have an organized proposal but sort of the kind of agenda that he has, as well as Arlen Specter's actual immigration proposal, envision unlimited guest worker flows. In other words, there would be no numerical caps. And one of the things that might affect and I'm not sure what the answer is, but might affect the level of education of the guest worker flow is that once it seems to me it's at least possible that once such a guest worker program were systematized and organized and no longer below ground like the illegal immigrant flow now, we might end up with huge increases in workers from, say, Pakistan and Indonesia and Bangladesh really cheap labor countries, because Mexico is an upper-middle-income country by world standards. It's already priced itself out of the labor market. Factories are closing in Mexico and moving to China.
So the point is that we could end up with a conceivably anyway with a much lower level of education among the flow of people coming in from outside even than we have now if these guest worker programs are instituted and institutionalized.
Yes, sir? Yeah, you.
Q: I'm Al Milliken (sp), Washington Independent Writers. At these employment locations like the 7-11 in Virginia that was mentioned and I'm just thinking of my knowledge in D.C. would you identify a place like 15th and P it seems to me that that kind of activity goes on there too employment, or attempt at getting employment. Do we know why this is populated by what you're identifying as Central American young males? Do we see any comparable employment place for unemployed or underemployed native Americans to gather, for example, outside homeless shelters or soup kitchens? And do we have knowledge about what law enforcement or lack of policing is done at these mass gathering places?
MR. HARRINGTON: Well, it's an interesting question. I don't know about that P Street location. I can tell you there have been some efforts to actually the size of these things can get quite large; there will be 200, 300 men standing on a corner. And oftentimes what you'll see happen in different parts of the country is the retail trade organizations in that area will actually organize themselves, get a county or state grant to actually get a facility to bring these things indoors, and then they try to regulate them. They try and get firms to pay kind of a standard wage, and the like. What we find is sometimes that when we bring these things indoors to try and regulate them, organize them a little bit, then a few months later a new outdoor shape-up will get organized.
Are there surveys of these things? The answer is no. Traveling around the country I've just seen the growth in these kinds of organizations. These things, just as far as I can see, just didn't exist probably even 15 years ago.
MR. BERNSTEIN: There are two surveys the workers centers is one of the things that Paul is describing and we have one on our website, EPI.org, and it's by Janice Fine, and it's a new book which surveys workers centers and provides you with lots of information about them that you might find interesting. I don't know the answer to your I mean, the other part of your question about homeless shelters, I don't know of such organizations, and it's an interesting difference that you raise.
MR. KRIKORIAN: If I could just respond to something Jared had said earlier, is there this is sort of for whoever wants to take it what is the sort of common ground that you guys have? In other words, what are the areas of agreement, because you were highlighting, and I think importantly, areas and caveats and all that, but what are the areas of accord?
MR. BERNSTEIN: I'm glad you asked that. Let me say, one, there are no labor shortages in the low-wage sector. We all agree with that. And, you know, you just look at the wage trends there and I don't see how you could possibly make an argument for a labor shortage.
MR. CAMAROTA: Okay, I was going to ask that maybe put it that way, that America does seem to have a very large supply of less-educated workers. We all agree on that, that they have not fared well in the last five years, and that is very strong prima facie evidence that that kind of labor is not in short supply.
MR. BERNSTEIN: Neither has the high end.
MR. CAMAROTA: Yeah.
MR. BERNSTEIN: The wage of college graduates fell about 5 percent in real terms 2000 to 2004.
MR. CAMAROTA: Yeah, it's interesting, but again, in terms of the illegal immigration debate or the debate about immigration reform right now, it generally focuses on the bottom, but it's a fair point. In general there just doesn't seem to be that kind of so there's a lot of less-educated natives. They haven't done well in the last five years, and the idea that there are jobs that only immigrants do doesn't make any sense. The economy is not structured in that way. The vast majority of almost every single occupation in the United States is done by natives. Where we get a disagreement, I think, right, is how big an impact is immigration likely to have? You say some but not sure, and I think that I come down and I don't want to speak for Paul is that, no, it's probably bigger than some of your comments have suggested based on, I think, who is being affected and so forth. Is that would that be kind of fair?
MR. BERNSTEIN: I think that's fair. I mean, the only addition I would add to you know, this thing that we all agree that there this notion there are jobs Americans won't do is really, I think, a pretty destructive notion in this debate. I will say there are jobs that if you degrade the quality of those jobs such that only the most desperate person will take them then, yeah, you're going to fill those jobs with people who are desperate for work, and that's, I think, more what we're looking at. But if the quality of those jobs were to improve significantly, or course other people would take them.
MR. KRIKORIAN: Yes?
Q: I'm John Wahala, also with the center. How does the premise of immigrants shape the development of an occupation or industry? You guys have talked about the impact on benefits and wages. Well, what about working conditions and the adoption of technologies that might make the job safer or more easier to do?
MR. KRIKORIAN: I have some thoughts on that. It seems to be pretty clear that in those industries where low-skilled labor from overseas is concentrated and increasing, that the adoption of new techniques of mechanization, et cetera, is actually slowed down, and vice versa. When foreign labor is reduced or constricted to those industries or occupations, they actually pretty rapidly become more productive, more mechanized, what have you. I mean, the most extreme cases are agriculture where and actually, this was the lead of Samuelson's column today, that when the previous guest worker program we had that proponents are now somehow pointing to as a model for what we should do in the future when that was ended, the importation of Mexican farm workers was ended in the early 60s, that the tomato industry in California, processing tomatoes for sauce and juice and what have you, which very heavily relied on this labor essentially the employers concluded that their investment in lobbyists had failed so they started to invest in mechanization, in perfecting tomato harvesters, in developing a variety of tomato that was more consistent with being picked and processed by machine, and production increased massively and the real post-inflation price of the tomato products fell because the industry modernized.
Likewise with sugar. In south Florida 20 years ago it was all harvested like it was the Middle Ages, with sweaty, shirtless men hacking away at plants with machetes, and Americans weren't going to do that kind of work, as Jared points out. The conditions were in fact so bad that Americans weren't doing it, so Jamaican guest workers were brought in. But their contracts were violated repeatedly by the farmers, and that gave rise to lawsuits, a whole wave of lawsuits, which raised the price of that labor. Farmers invested in sugar harvesters; 100 percent of sugar is now harvested by machine.
So clearly, in the long run even the industries that are using this cheap labor are harmed because their development is distorted. They're competing with overseas competitors on the basis of their competitive advantage cheap labor whether it's Turkish raisin farmers or anything else rather than our competitive advantage, which is capital and technology.
MR. CAMAROTA: Let me add one important part of that story with the tomato farming as well. The best part of that story is what happened when we ended the guest worker program with Mexico. Wages went up 40 percent for American farm workers. There were fewer of them, they were more productive, and they made a heck of a lot more money. The job was civilized. And that, I think, is important. A guest worker program, by increasing the supply of labor, was clearly I mean, everyone who has looked at this shows that it was almost certainly holding down wages. And the beauty of that was that some of the poorest American workers now did better with the end of the guest worker program. And, I should add, as Mark has pointed out, because of mechanization it was completely transparent to the consumer.
Labor costs in unskilled labor are only a tiny fraction of economic output anyway. We didn't see any spike in the price of tomatoes anyway because they mechanized. And as I say, maybe the best part of it is the farm workers did a whole lot better. Now, when illegal immigration kicked up again and the supply of labor was increased, a lot of that progress was unfortunately undone.
MR. BERNSTEIN: Just one quick comment. I mean, Mark and Steve spend their live studying immigration and know it cold, and there is a lot to what they say, but I guess I try one thing I hope I can contribute to this is a somewhat broader view. I don't disagree with that kind of industry-level analysis, but if you look at the 1990s, here was a period of large historically large immigrant inflows. Productivity nationally I'm not talking about a particular industry productivity for the nation accelerated it's called the productivity miracle. It went from growing about 1 percent a year to between 2 , 3 percent per year. That extra 1 percent is a huge contribution to American living standards. At the same time, the wages of both immigrants and natives rose with productivity for the first time in 30 years.
So manipulating immigrant flows can have the kinds of outcomes they're talking about I'm not denying that at all but there are other ways through which living standards and productivity can be boosted that are not related to immigration.
MR. KRIKORIAN: Well, let's take another one or two questions. Yes?
Q: I wanted to ask also this kind of culturalization. I've had two incidents which make me feel that as a concentration of particular cultures get into an industry, it does affect that. One is an African American who came to my house looking for work. It turns out he was a construction worker, said he couldn't get a job in D.C. because he didn't speak Spanish in the construction industry. At the other end of the spectrum, I have a friend who was accepted into an MIT Ph.D. program in computer engineering, decided not to do it because he said he didn't want to spend 24/7 in a lab where everyone spoke Chinese.
So I'm wondering if a heavy concentration of particular immigrant groups go into particular labor forces, if that actually discourages Americans from going into that
MR. CAMAROTA: Right, we talked about wages and benefits and working conditions, and we mentioned some other reasons why immigrants might crowd out natives. This is an interesting point, that if you don't speak one of the things that's happened with U.S. immigration is in one way it's become much less diverse. People who study it talk about this. We do get people from all over world in a way that we didn't. On the other hand, one linguistic cultural group has come to dominate new immigration, both legal and illegal, with 50 percent-plus of all new immigrants speaking only one language. We don't seem to have had that before in our history. As a consequence, if you're in the labor markets where a lot of those immigrants are concentrated and you don't speak that language, you might well be at a significant disadvantage because your coworkers can't communicate with you and your employer is less likely to promote you to foreman if he doesn't think that you so that is another interesting possibility and it's kind of one the consequences of very large-scale immigration that is not so diverse as it used to be.
MR. BERNSTEIN: Do those anecdotes contradict your view about lack of concentration that you stressed?
MR. CAMAROTA: Well, I mean, if it's 15 percent foreign-born in the occupation, a person who comes in and says, I speak both languages, you would suspect that the employer would say, hey, all right; but if someone comes in and says, you know, I don't speak that other language, he might be at a disadvantage even though the overwhelming percentage of the workers are native born. Even if it's just one out of eight or one out of nine, that's still a pretty nice thing because the employer's going, gosh, I can't speak to any of them, I've got a problem, but now I've got you; come on in.
MR. KRIKORIAN: And likely it's going to be more concentrated in certain regions as well.
MR. CAMAROTA: Right.
MR. KRIKORIAN: Let's take one more question if there is one and then okay, then you get a second bite at the a quick question.
Q: Al Milliken again. I've seen your early report issued in December 2005. You claim that one-third of immigrants lack health insurance. When an immigrant is seriously hurt so they're not able to work, what happens to that worker, and what responsibility do we know that employers take? I do know of an injured
MR. KRIKORIAN: Let's let them answer the question. We're going to run out of time.
MR. CAMAROTA: Yeah, I guess two quick points. One of the down sides of letting in a lot of unskilled people, putting aside labor market competition, is that there is a consensus in the immigration literature that people with relatively little education pay relatively little in taxes but tend to use a lot in services, like health care. The National Academy of Science has estimated that in the course of his lifetime, an immigrant without a high school degree would use $89,000 more in services than he paid in taxes, and an immigrant who had only a high school degree would use $31,000 more in services than he paid in taxes. However, more educated immigrants were a significant fiscal benefit. And so this is very germane to the question of letting illegal aliens who are overwhelmingly unskilled stay and bringing in even more unskilled people. It's different than the labor market issues.
Employers generally strive to maximize their use of immigrant labor and socialize the costs. They try obviously they're not providing health insurance that's good. They like to try to have a situation where they get a state legislature to pass rules that say if someone is injured on the job and it turns out he's illegal, I don't have to pay benefits to that person. That seems very reprehensible but employers have pushed for that kind of thing. It's consistent with this idea of maximizing the individual benefit for the employer and socializing spreading out the costs to taxpayers. And certainly we should not let that sort of thing happen.
MR. KRIKORIAN: Well, thank you, Steve. Let's wrap it up here. I know at least some of us will be up here for those of you who still want to accost us with other questions; I'm not sure if all the panelists will. I appreciate your coming. The report is online at CIS.org. And thank you very much.