Borjas gives us both (i) a good synopsis of some of the "facts" of immigration that economists ascertain and work with, as well as (ii) valuable insights into the limits of economic analysis. While his account strikes me as honest and helpful, what resonates particularly strongly with me is the extent of ignorance that we face in looking at a phenomenon like immigration - something that comes out more graphically if you actually watch Borjas' lecture which is embedded at the bottom of this post. In the below text, however, I try to distil as much incontrovertible information as possible.
Even though we know so little for sure, "everyone" takes her position in the matter as if knowledge problems were trivial. What this suggests to me is the importance of the institutions of freedom in enabling a peaceful and socially non-disruptive debate and policy making process among uninformed and highly antagonistic members of society.
I do not think we can account for such relatively civilised conflict resolution without a theory of how political action triggers not only
- primary effects - its intended procedures and aims -
- secondary effects that operate behind our backs as if moved by an invisible hand - the invisible hand of politics and the state.
This is ultimately, what I am striving for in looking at immigration: to better understand the part played by the invisible hand of politics. I hasten to add that I do not expect to find a self-controlling mechanism, but an order partly grown and partly designed where spontaneous processes amalgamate intricately with planned measures.
Let us now turn to some economic insights pertaining to immigration in the US:
Summary of Borjas' Lecture
Borjas' lecture starts at around time-mark 05:00 (see the embedded video at the bottom of the post).
The two core issues of immigration in Borjas view:
Time mark 07--30: How many immigrants is a host country willing to admit? (Numerical limit)
08--30: Which immigrants to you want to let in - there are billions to choose from? (Allocation system for limited number of visas)
09--00: These two core issues have never been addressed, at least since 1990.
09--15: The lecture is about looking at answers to these fundamental questions which one may garner from economic theory and empirical research, answers that we may hope allow us to replace sentiment and passion with fact-based rational insight.
10--00: 13,5% of US population was foreign born in 2010, quite in sync with similar percentages in other countries of the industrialised world.
11--45: Two peaks of US legal immigration (by decade), one around 1900, the other around 2000.
11--56: Today, roughly 1 million people entering illegally per year.
12--10: Four historical stages of US immigration policy:
- Before 1875 - no restrictions
- 1875-1924 - defined "excludables," comprising Asians, public charges etc.
- 1924-1965 - national origins quotas (first time application of a numerical limit & allocation by extant proportions of national extraction (if 5% of US pop. Italians, then future eligibles again 5% Italians)
- Since 1965 - family preference system (in the spirit of the civil rights movement - allocation privileging people with family connections in the US)
15--30: Contemporary classes of admission (2001 - 2010):
- Legal immigrants (total) - 10.5 Mio., of which
- Family-preference - 6.8 Mio.
- Employment-based - 1.6 Mio (including dependants)
- Refugees/asylees - 1.3 Mio
- Lottery winners - 453 thousand
17--40: Illegal Immigrants (25% in California, 60% Mexicans)
- 2000 - 8.5 Mio.
- 2005 - 10.5 Mio.
- 2007 - 11.8 Mio.
- 2008 - 11.6 Mio.
- 2010 - 10.8 Mio
- 2011 - 11.5 Mio.
There is a big debate about these figures that underlines the enormous number of uncertainties lurking behind the picture we try to draw of immigration reality.
Even where data ("facts") are available, they often do not easily make for a consensus on the facts. That is before we turn to issues related to subjective perceptions and convictions.
20--50: Why we have an immigration debate in the US - zero sum assumption of economic impact.
Percentage wage gap between immigrants and native men (age-adjusted) declining from
- 1960 - + 5%
- 1970 - +- 0%
- 1989 - - 12%
- 1990 - -14%
- 2000 - -18%, to
- 2010 - -22%
"The fact that there has been a decline in economic performance of immigrants ... really underlies most of the questions at the core what we care about politically."
22--50: Who are the immigrants? What does economics tell us about the reasons why only relatively few immigrate (10% of all Mexicans), while the majority do not emigrate, even in the face of free movement, as in the case of Puerto Rico?
24---00: Countries with a narrow range of income distribution, like Sweden, the highly skilled have a strong incentive to emigrate, as the returns to skills are relatively low (a doctor not making much more than a bus driver); whereas with a wide distribution of income (the rich being very rich and poor very poor - and high returns to skills), the low-skilled have strong motives to emigrate, as they stand to make palpable gains in income.
25--00: Immigrants from countries with a narrow income distribution (such as Sweden) tend to perform economically much better in the US than those from countries with a high Gini coefficient (indicator of income inequality).
25--38: Another strong correlation (between economic performance of immigrants and GDP of the source country) confirms:
"Clearly, people who come from wealthier countries have skills that ... tend to be more easily transferable to the US."
26--12: Do immigrants alter the employment opportunities of natives? Talking at a time when strongly limited and discriminatory immigration policy was not contested politically, Paul Samuelson argued in 1964 - just before the 1965 change in US immigration policy:
"By keeping labour supply, down, [severely restrictive] immigration policy tends to keep wages high."
28--00: Yet again (as mentioned at 17--40), it turns out incredibly difficult to corroborate the facts (?) that seem so suggestive when applying straightforward supply-and-demand analysis. One factor complicating the issue is the dispersal of non-comparable labour markets (school drop outs in their 20s, graduates in their 40s etc.). And this includes only two differentiating criteria: education and age. Also these groups change their economic profile over the years, while new entrants appear. Trying to accommodate these factors, Borjas arrives at the following conclusion:
For whatever reason, over the last 50 years, the groups that experience the highest influx of immigrants were the groups that experienced the slowest wage growth in that decade.
31--40: It works also the other way around, when you take out 10% members of a certain skill group (as happens thanks to emigration in Mexico or Puerto Rico, the wage levels increases.
37--45: Who gains? Who loses?
Borjas has the below figures and feels that there is distributive conflict at the bottom of immigration:
- Workers lose 2.8% of GDP, or 400 billion
- Employers gain 3 % of GDP, or 430 billion
- Net gain is about 30 billion annually, about $ 110 per native-born person.
Flows of wealth are much greater than the net gain.
39--50: Finally, what is the fiscal impact of immigration - an American concern since 1645?
42--10: In contrast to Julian Simon - see Immigration and Freedom (1/10a)-, Borjas seems to be able to identify considerably stronger reliance of immigrants on public assistance compared to natives, implying a significant fiscal impact.
In view of the above finding, what should we do about immigration policy?
44--00: NOTHING AT ALL!
At least, as long as we cannot answer this question: WHAT DO WE WANT TO BE AS A COUNTRY? What do we want to accomplish by an immigration policy?
44-33: To the humanitarian, Borjas says:
The US, right now, has the biggest anti-poverty programme ever run in the entire universe through its immigration policy.
Low-skill worker wages are negatively impacted and citizens may have to pay more taxes, but if one's goal is to reduce poverty worldwide, these costs may appear worth incurring.
Putting on the hat of the tax payer, who only cares about his income and his retirement, Borjas, wonders why should I make sacrifices to support a 33% dependency rate in immigrants?
Same facts, but different objective functions, lead to very different immigration policies.
Man cannot live by facts alone [if we can bring ourselves to share common facts]. Qua economist, one is not in a position to recommends policies. Before one can take a meaningful stand on immigration, one needs to insert additional substantive content (representing, for instance, humanitarian ambition or taxpayer interests) whose determination is not the business of the economist.
Be this as it may, countries do tend to erect admission hurdles, frequently by using a point system, which in the US is implicitly operative by the fact that people with US family connections ipso facto score heavily, while Canada screens with its point system for likelihood of economic success.
Establishing a point system raises the core issue as to what it is that you wish to attain with an immigration policy, which objectives do you wish to foster?
Policy implications do not come from facts alone ... Whose well-being do you want to maximise? What do you care about?
It is very difficult to come up with an immigration policy that makes all three groups better off (natives, immigrants, and those that are left behind) ... You have to allocate weights to different groups. The allocation of weights is what the objective function is about and what in the end the immigration policy is about.
Yet, attaching weights to specific groups is a practice that conflicts with cultural preferences that have become dominant in Western countries.
One might surmise that in a free society, we will tend to encounter a policy mix geared among other things to disparate types of policy-consumers, each policy signalling compliance with the specific demands of a targeted group influential enough to receive placating signals.
To the extent that contradictory or incoherent policies have detrimental effects, these must be counted as costs of liberty.
The costs of liberty are the costs generated by a regime that
- allows dissent on all (including explosive) issues, and therefore simultaneously
- faces a heightened need to safeguard social cohesion
in a social order whose overall success hinges vitally on accommodating a highly disputatious community with relatively low barriers to political influence by any citizen ambitious to make an impact.