VDARE - Time to Rethink Immigration? (PART 2)
Time to Rethink Immigration? (PART 2)
Mr. Brimelow is Editor at VDARE.com.
Click here to return to Part 1...
Asking the Right Questions
Asking the Right Questions
SUPPOSING AMERICA'S political elite suddenly decided to notice immigration, what questions should they consider?
Is immigration really necessary to the economy?
Audiences always burst out laughing at one apparently gagless scene in the hit movie Back to the Future: the time-transported hero drives up to a gas station in the 1950s, and an army of uniformed attendants leaps forth to pump the gas, clean his windshield, fill his tires, polish his hubcaps, offer him maps, and so on. The joke was in the shock of self-recognition. It was only yesterday—and yet completely forgotten, so accustomed is everyone now to self-service.
"We need immigrants to meet the looming labor shortage/do the dirty work Americans won't do." This further item from the pro-immigration catechism seems to be particularly resonant for the American conservative movement, deeply influenced by libertarian ideas and open, somewhat, to the concerns of business.
But it has always seemed incongruous, given persistent high levels of unemployment among some American-born groups. Since these groups obviously eat, it would appear that public policy is subsidizing their choosiness about work, thus artificially stimulating the demand for immigrants.
And if there is a looming labor shortage (hotly disputed), it could presumably be countered by natalist policies—encouraging Americans to step up their below-replacement birthrate. Even the current high immigration inflow is exceeded by the 1.6 million abortions in the U.S. each year.
For example, the federal income-tax code could be adjusted to increase the child allowance. In 1950, this provision exempted the equivalent (in 1992 dollars) of $7,800 for each child; now, after inflation, it exempts only $2,100. Or the "marriage penalty"—by which a couple pay more in taxes if they marry than if they live together out of wedlock-could be abolished. Or the public-school cartel could be broken up, reducing the crushing costs of educating a child.
But Back to the Future makes a more fundamental point: labor is not an absolute. Free economies are infinitely ingenious at finding methods, and machinery, to economize on labor or any other scarce resource.
The implicit assumption behind the economic argument for immigration appears to be something like this:
Labor x Capital = Economic Growth
So, for any given capital stock, any increase in labor (putting aside the question of its quality) will result in at least some increase in output.
This assumption is just wrong. Typically, technical studies that attempt to account for economic growth find that increases in labor and capital account for at most half and often much less of increases in output. Simon Kuznets's survey of the growth of the West over the last two centuries concluded that increases in labor and physical capital together were responsible for less than 10 per cent of the greatest output surge in human history. The rest seems to be attributable to changes in organization—to technological progress and ideas. Or:
Economic Growth = Labor x Capital x {???}
And {???} is dominant.
Back to the Future illustrates this process in action. On the face of it, gas stations have simply substituted capital (the self-service pumps) for labor (gas jockeys). But actually what has happened is more complex: the cost of making the pumps, and of designing the computer system behind them, is far exceeded by the savings on labor, which extend indefinitely into the future. It is reorganization that has resulted in a permanent increase in productivity.
From an economist's standpoint, the factors of production are not absolutes, but a fluid series of conditional interacting relationships. This insight won Julian Simon one of the famous debating victories of our age. In 1980, he bet the well-known liberal doomster Paul Ehrlich that several commodities Ehrlich claimed were running out would in fact be lower in price in 1990, the economy having adjusted in the meantime. They were, and Ehrlich had to pay up. Paradoxically, however, when it comes to immigration, Simon seems to revert to a classic non-economic view: Labor is good, more labor is better.
The economic view of labor has influenced the current immigration debate only in one direction: it is triumphantly produced by the pro-immigration side to refute any unwary critic of immigration who assumes that native-born workers must inevitably be displaced. They aren't, necessarily, in aggregate, because the economy adjusts; and because the increase in the factors of production tends to create new opportunities. "Immigrants not only take jobs," writes Julian Simon, "they make jobs."
Maybe. But missing from the current immigration debate is the fact that this effect operates in the other direction too. On the margin, the economy is probably just as capable of getting along with less labor. Within quite wide boundaries, any change in the labor supply can be swamped by the much larger influence of innovation and technological change.
The historical importance of immigration to the U.S. can be exaggerated. Surprising as it may seem, demographers agree that the American population would be about half its present size that is, much bigger than Germany's and about as big as Japan's-even if there had been no immigration after 1790. Even more significantly, the Harvard Encyclopedia of American Ethnic Groups estimates that immigration did not increase U.S. per-capita output at all. Indeed, both France and Germany outstripped the U.S. in growth of per-capita output in the hundred years after the mid nineteenth century.
Absolute size can be useful while seizing a continent or fighting wars. But in the end it is output per capita that determines living standards. And both proportionately and absolutely, in an increasingly technical age, what will count is not the quantity of people but their quality—and the quality of their ideas.
The {???} factor is the explanation for the great counter-factual episode hanging like the sword of Damocles over contemporary pro-immigration polemics: the success of Japan since World War II. Despite its population of only 125 million and virtually no immigration at all, Japan has grown into the second-largest economy on earth. The Japanese seem to have been able to substitute capital for labor, in the shape of factory robots. And they have apparently steadily reconfigured their economy, concentrating on high value-added production, exporting low-skilled jobs to factories in nearby cheap-labor countries rather than importing the low-skilled labor to Japan.
It is highly significant of the false nature of the American immigration debate that, despite all the public hysteria about Japan, no attempt is ever made to look for lessons in its immigration policy. Incredibly, although his book is called The Economic Consequences of Immigration, Julian Simon simply ignores the subject altogether. Asked about it by Forbes magazine's Jim Cook, he in effect struck out: "How Japan gets along I don't know. But we may have to recognize that some countries are sui generi s."
However, Simon's view of the impact of immigrants does include important qualifications, which his enthusiastic acolytes often miss. Simon believes that native-born workers are not necessarily displaced in aggregate. In his book, he frankly and repeatedly acknowledges that "Any labor-force change causes some groups to suffer some harm in the short run... It is true that some particular groups may be injured by a particular group of immigrants ..." (This works in reverse. Agribusiness lobbies for cheap immigrant labor rather than mechanize itself, regardless of the overall cost to the economy. Ironically, agribusiness is itself often subsidized—for example, by federal water projects.)
As it happens, the U.S. contains one particular group that is clearly vulnerable to competition from immigration: blacks. This question has attracted attention for years. Immigration from Europe after the Civil War is sometimes said to have fatally retarded the economic integration of the freed slaves. Conversely, no less an authority than Simon Kuznets felt that the Great Immigration Lull after the 1920s enabled Southern blacks to begin their historic migration to the cities and the economic opportunities of the North.
Blacks themselves take a dim view of immigration, according to opinion polls. In the FAIR poll cited above, 83 per cent of blacks thought Congress should curb immigration. But George Borjas found that blacks living in areas of immigrant concentration did not appear to have suffered significantly reduced incomes compared with those elsewhere. The reason, he theorizes, is that during the years in question—the 1970s—the effect of immigration was overwhelmed by the effects of baby-boomers and women entering the labor market. Now, of course, these factors no longer apply. Additionally, studies of high-immigrant areas may fail to capture a tendency for native-born workers to relocate because of the increased competition. Across the entire country, the wages of native high-school dropouts fell by 10 per cent in the 1980s relative to the wages of more educated workers. Borjas calculates that about a third of that decline is attributable to immigration.
Borjas, moreover, was perturbed by the tendency of low-skilled recent immigrants, not necessarily to displace American blacks, but to join them in swelling the ranks of the underclass: "Few issues facing the U.S. are as important, and as difficult to resolve, as the persistent problem of poverty in our midst... The empirical evidence presented here suggests that immigration is exacerbating this problem."
Since the Great Society, a significant part of the black community has succumbed to social pathology. There is at least a possibility that this is related to the simultaneous opening of the immigration floodgates. In which case, it is perhaps to current policy, and not to critics of immigration, that the over-used epithet "racist" might best be applied.
Another important Simon qualification, unnoticed by his acolytes, is his concept of "negative human-capital externalities." Most recent immigrants have lower skill levels than natives, he notes. If enough of them were to arrive, they could overwhelm and render less effective the higher skills of the natives. "In other words, if there is a huge flood of immigrants from Backwardia to Richonia, Richonia will become economically similar to Backwardia, with loss to Richonians and little gain to immigrants from Backwardia ... So even if some immigrants are beneficial, a very large number coming from poorer countries ... may have the opposite effect."
This is a crucial theoretical concession. Coupled with the fact that the numbers and type of potential immigrants are unknown, it is the reason Simon quietly declines to follow the logic of his other arguments and endorse completely open borders (as, for example, the Wall Street Journal editorial page has done). Of course, he insists that immigration levels could be much higher than at present. But Richonians in California, Florida, and New York City might not agree.
"You have to accept the free movement of people if you believe in free trade/free markets." You do? It's a more radical proposition than appears at first sight. Third World populations are very large and their wage levels very low—Mexican wages are a tenth of those north of the border, and Mexico is relatively advanced. So calculations of the market-clearing wage in a U.S. with open borders necessarily imply that it must be some fraction of its present level. This arrangement might optimize global economic utility. But it can hardly improve American social harmony.
However, a calculation of this sort requires impossible assumptions. The fact is that a belief in free markets does not commit you to free immigration. The two are quite distinct. Even Julian Simon, although he favors immigration, says explicitly that immigration's benefits are not from "trade-like effects":
Contrary to intuition, the theory of the international trade of goods is quite inapplicable to the international movement of persons. There is no immediate large consumer benefit from the movement of persons that is analogous to the international exchange of goods, because the structure of supply is not changed in the two countries as a whole, as it is when trade induces specialization in production ... the shifts due to international migration benefit only the migrant.
On a practical level, free trade actually tends to operate as a substitute for immigration. Hence the Japanese have factories in the Philippines rather than Filipinos in Japan. And Victorian Britain, with its grand strategy of "splendid isolation" from the quarrels of Europe, combined total free trade with almost no immigration, a policy that satisfied Liberal "Little Englanders" and Tory Imperialists alike.
In theory, free trade with Mexico should help reduce the current immigrant flood by providing work south of the border. In practice, however, "free-trade negotiations" (a paradox: what's to negotiate?) often get captured by political elites seeking to favor client constituencies. Rumors that the current talks with Mexico might lead, absurdly, to an increase in immigration suggest this insidious process is well under way.
A commitment to free trade and free markets does not mean that you would sell your mother if the price was right. The free market necessarily exists within a social framework. And it can function only if the institutions in that framework are appropriate. For example, a defined system of private property is now widely agreed to be one essential precondition. Economists have a word for these preconditions: the "metamarket." Some degree of ethnic and cultural coherence may be among them. Thus immigration may be a metamarket issue.
At the very least, a diverse population increases what in economics-speak are called "transaction costs." Dealing with people whom you don't know and therefore can't trust requires expensive precautions. I suspect this is one factor behind the legalism infesting business practices in the U.S., as compared to Britain. Beyond this, capitalism generates inequality and therefore envy. And such emotions can be much more intense across ethnic and racial lines—witness the fate of the Korean storekeepers in Los Angeles.
This is not an unprecedented insight. Friedrich von Hayek, the first classical liberal to win the Nobel Prize for economics, used to advance a sort of sociobiological argument for the apparently immortal appeal of socialism. Cities and civilization have come very late in human history, he pointed out. Almost all mankind's experience has been in small hunter-gatherer bands. Face-to-face relationships are still much more comprehensible to us than impersonal ones. Thus an increase in rent is blamed on the greed and obnoxiousness of the individual landlord, and provokes an irresistible urge to bash him with rent controls, despite all the evidence that this leads merely to shortages and inequity. And, to extend Hayek's argument, it is obviously easier to demonize a landlord if his features are visibly alien.
Another classical liberal Nobel Laureate, Milton Friedman, has speculated that the culture of the English-speaking world itself may be, from an economic standpoint, sui generis . . . in Simon's phrase. I interviewed him for Forbes magazine in 1988:
FRIEDMAN:... The history of the world is the history of tyranny and misery and stagnation. Periods of growth are exceptional, very exceptional.
BRIMELOW: You've mentioned what you see as the institutional prerequisites for capitalism. Do you think there might be cultural prerequisites too?
FRIEDMAN: Oh, yes. For example, truthfulness. The success of Lebanon as a commercial entrepot was to a significant degree because the merchants' word could be trusted.
It cut down transaction costs.
It's a curious fact that capitalism developed and has really come to fruition in the English-speaking world. It hasn't really made the same progress even in Europe—certainly not in France, for instance. I don't know why this is so, but the fact has to be admitted.
Eschewing these more subtle considerations, George Borjas focuses on the quantifiable. His conclusion is stark. "The economic arguments for immigration simply aren't decisive," he told me recently. "You have to make a political case—for example, does the U.S. have to take Mexican immigrants to provide a safety valve and keep Mexico stable?"
Put it another way: for the U.S., immigration is not an economic necessity. It is a luxury. Like all luxuries, it can help-or it can hurt.
Is immigration really beneficial to society?
Forty-four years ago, Richard Weaver published a book the title of which, at least, convinced the conservative movement: Ideas Have Consequences . It is now time to recognize a further truth: Immigration Has Consequences.
The crudest consequences relate to political power. Because many libertarians and economic-growth conservatives are so reluctant to admit this logical possibility, it is worth emphasizing that there are plenty of examples of immigrants and their descendants threatening the political balance of a state (polity), from the Uitlanders in the nineteenth-century Boer Republics to the Indian politicians recently elected to govern Fiji and promptly deposed by the ethnically Fijian army. And how about this chilling comment from the Harvard Encyclopedia?
In obtaining land grants in Texas, Anglo immigrants agreed to become Mexican citizens, obey Mexican laws, accept the official Catholic faith, learn Spanish, and take other steps to become fully assimilated as law-abiding citizens. However, over the years, it became clear that these settlers, now Anglo-Mexicans, were not becoming integrated into the nation and that Anglo immigration had become a problem . . . The strains and disagreements ultimately led to the Texas Revolution in 1835.
Er, quite.
These political consequences need not threaten the integrity of the state (polity)—just its foreign policy.
Thus domestic ethnic-group pressure clearly plays a role in Washington's essentially contradictory attitudes to the white settler communities of southern Africa and the Middle East.
But probably the most important consequences are cultural. "The most obvious fact about the history of racial and ethnic groups," writes Thomas Sowell in The Economics and Politics of Race, "is how different they have been—and still are." Sowell's work, carried on in Ethnic America: A History, conclusively demonstrates that cultural patterns are pervasive, powerful, and remarkably persistent, even after generations of living under common institutions, as in the United States. (Similarly, David Hackett Fischer's monumental Albion's Seed recently traced America's dominant folkways all the way back to four distinct waves of colonial immigration from different regions of Britain.)
"But aren't these consequences good?" Naturally, there isn't anything in the pro-immigration script about cultural consequences. However, this is the usual reaction if you insist on raising the point. It's embarrassing, of course. In the current climate, it is impossible to discuss the failings of any ethnic group.
But look at it this way: Thomas Sowell's work shows that cultural traits, such as attitudes to work and education, are intrinsically related to economic success. Germans, Japanese, and Jews are successful wherever they are in the world. Conversely, the work of George Borjas and others shows that national origin, a proxy for culture, is an excellent predictor of economic failure, as measured by propensity to go on welfare. In a recent paper, Borjas has demonstrated that disparities among the 1880-to-1920 immigrant groups have persisted for as much as four generations. Thus there can be absolutely no question that the cultural characteristics of current immigrant groups will have consequences for the U.S.—in this case, economic consequences—far into the future.
The same argument applies to crime. Random street crime, the great scandal of American cities since the 1960s, is clearly related to impulsiveness and present orientation, a key cultural variable. More significant, however, is organized crime. This has typically been ethnically based, partly because it reduces the criminals' transaction costs and because such groups are difficult to penetrate.
In recent years the Mafia or Cosa Nostra has been in decline, not least because of the acculturation of Italian-Americans. But this is "dirty work" that some of the post-1965 immigrant groups are positively anxious to do-more violently, particularly in the burgeoning drug business, than the Mafia ever was. There are several such new "mafias," staffed by Russian Jews, Hong Kong Chinese, Colombians, and even less well-known communities like the Chaldeans—Iraqi Christians whose convenience stores in the Detroit ghetto are centers of criminal activity.
Today such news would be judged unfit to print regardless of its accuracy. Researchers find that official figures on immigrant and ethnic crime patterns are rarely collected. That certain ethnic cultures are more crime-prone than others, however, must be considered a real possibility.
Curiously, Congress appears to have shaken off its general paralysis to recognize that immigration can have cultural consequences—for Pacific Islanders. Five U.S. territories, American Samoa, Micronesia, the Marshall Islands, the Northern Marianas, and Palau, have been given control over immigration to protect their ethnic majorities. In American Samoa and the Northern Marianas, U.S. citizens cannot even own land unless they are Samoan, Chamorro, or Carolinian.
This double standard has incensed an extremely erudite and energetic professional writer in Rye, New York, Joseph E. Fallon. Fallon argues that controlling immigration is simply a question of American self-determination. And he is attempting to organize a class-action law suit challenging current policy on the grounds of the 1948 Genocide Convention, which banned "deliberately inflicting upon a [national] group conditions of life calculated to bring about its physical destruction in whole or in part."
Which, after all, is no crazier than much liberal litigation.
Is immigration really good for (ahem) the Republicans?
The fate of the Republican Party may not be of concern to the political elite as a whole. But it should worry those aspiring members of the elite who also consider themselves conservatives.
Ethnicity is destiny in American politics. This point was made definitively in Kevin Phillips's brilliant The Emerging Republican Majority (1968), which demonstrated that ethnic settlement patterns had an amazingly persistent influence on voting patterns. Phillips predicted on the basis of demography that the Republicans would replace the Democrats as the majority party. And he was undeniably right in the presidential contest, even if timid and unimaginative leadership has squandered the opportunity on the congressional level.
As a glance around any of their meetings will tell you, the Republicans are the party of the American majority; the Democrats are the party of the American minorities. On its WASP foundation, the Republican Party has been able to add the children of each immigrant wave as they assimilate. This was the unmistakable subtext of the 1988 presidential election. With a Greek-American nominee, and implicitly anti-WASP attacks on George Bush's "preppie-ness," the Democrats hoped to hold the 1880-to-1920 immigrant wave. But they failed, just as nominating John F. Kennedy in 1960 did not prevent the continued defection of Irish-Americans.
The post-1965 immigrants, however, are overwhelmingly visible minorities. These are precisely the groups that the Republican Party has had the most difficulty recruiting. And, Jack Kemp please note, this is not necessarily a question of the Republicans' making nice, or nicer, to minorities. It may reflect the more divergent minorities' different values, and their more radical feeling of alienation from white American society. Current immigration policy is inexorably reinforcing Jesse Jackson's Rainbow Coalition.
The strained sound you hear is the conservative leadership whistling as they pass by the rainbow. Prohibited by the Bland Bargain from discussing the problem, they have indulged in a frenzy of wishful thinking. "We get quite a good vote from some Hispanic groups." Well, Hispanics are not quite as Democratic as blacks—that's a statistical impossibility—but the Republicans still face an uphill struggle. Even the much-lauded Cuban vote has actually been quite split, electing the likes of Claude Pepper and Dante Fascell to Congress. And Republican success with Hispanics, as with other minorities, is often at the expense of conservative principles. "West Indians are different." Some West Indians do appear to have been more economically successful than American blacks, although it must be said that nowadays part of their enterprise goes into drug "posses" and car-theft rings. However, the skill level of the post-1965 wave of West Indian immigrants has deteriorated sharply. Caribbean immigrants are now the most prone of all to welfare dependency. And anyway, the political consequences were always illusory. Shirley Chisholm and Stokely Carmichael are both of West Indian descent. "The Asians are small-business types, education-minded, family-oriented—they're natural Republicans." So were the Jews, and look how they vote—still overwhelmingly and outspokenly Democratic despite the best efforts of a brilliant generation of conservative Jewish intellectuals. And Hawaii, where Asians predominate, is a Democratic stronghold. The truth is that no one really knows how the Asians will vote. But since 1965 they have become a minority twice as large as the Jews, and potentially at least as influential.
Is immigration really good for the environment?
American liberalism has survived the loss of its traditional issue, economic management, by improvising new ones. And environmentalism is one of the most important, both because it particularly appeals to the vocal upper middle class and because it appears to necessitate an interventionist government. Yet the single biggest problem for the environment is the fact that the U.S. population, quite unusually in the developed world, is still growing quickly. Immigration is currently an unusually large factor in U.S. population growth.
Like the impact of immigration on native workers, the relationship between population and pollution is subtler than it looks. A primitive band of slash-and-burn agriculturalists can cause more devastation than a much larger community of modern ex-urbanites with sealed sewage systems and manicured horse farms.
But only within limits. Something has clearly got to give if the population of California grows from 20 million in 1970 to 60 million by 2020, which is Leon Bouvier's upper-limit projection. (His lower-limit projection: a mere 44 million. Phooey!) The fragile desert ecologies of the Southwest may not be utterly destroyed. But they must be transformed. California will cease to be the Golden State and become the Golden Subdivision.
This prospect is presumably anathema to true environmentalists, who value wilderness in itself. But although a few were active in rounding FAIR, most of the professional environmentalist community in Washington avoid the issue. Which is a measure of the extent to which they have been co-opted by the liberal establishment-just like the civil-rights lobby, which never voices the anti-immigration sentiments widespread among the b