Immigration, holidays and computer programmers

Immigration has been a big political issue for at least the last fifteen years. That covers William Hague’s “foreign land” speech and Michael Howard’s “Are You Thinking What We’re Thinking”, as well as David Blunkett’s comments about schools being “swamped”*. Immigration now beats the economy in YouGov’s regular poll of most important issues facing the country**. If you conduct a focus group in the UK on pretty much any issue, at some point, someone will cautiously link it to immigration – at which point most of the rest of the group will join in.

But… immigration has not been a decisive issue in any UK General Election during my lifetime.

Here’s one way of explaining what seems like a contradiction. In experiments, people are offered two seaside holidays. Spot A has average weather, average beaches, average nightlife and medium temperature sea. Spot B has great weather, great beaches, no nightlife and cool seas. If people are asked which one they would like to go on, they choose Spot B. It’s got great weather and great beaches, right?  But a funny thing happens when you put the question a different way: say you have reservations with both holidays but have to cancel one, which would you cancel? Again, people choose Spot B. It’s got no nightlife and you can’t swim there, after all. Somehow Spot B has become everyone’s favourite and least favourite holiday destination.

Immigration is a bit like Holiday Spot B. The same person can feel strongly about not liking people speaking another language on their train and believing that racists are dreadful people. What issue should the government listen to the people more on: immigration or the economy? Immigration. What issue should the government avoid inflaming populist passions on: immigration or the economy? Immigration. More reasons come easily to mind for immigration than for the economy, but those can be reasons to be anti-immigration or anti-anti-immigration.

Related to this is the idea of “evaluability”. A different experiment has people pretending to hire two computer programmers. Smith has written 70 computer programs in the language the company uses and has a Grade Point Average of 3 out of 5. Jones has written 10 computer programmes in that language and has a Grade Point Average of 4.9 out of 5.

The trick here is that Grade Point Average is easy to evaluate: Jones has 4.9 out of 5 (great), while Smith only has 3 out of 5 (middling). How much experience of computer programming is needed is much harder for the participants in the experiment to evaluate. So if the two candidates are examined separately, Jones gets paid more.

However, if you make the evaluation easier by looking at the two CVs together, the situation changes. This is a computer programming job so programming experience takes priority, and now it’s clear that Smith is considerably better than Jones: and Smith gets the higher salary.

Immigration is very easy to evaluate. Helpfully, when They come over Here, They are marked out by having different skin, different accents or, at the very least, funny names***. No one will ever be short of evidence that They (no, the other, They – government) have let immigrants in.

In contrast, the performance of the economy, the character of a Prime Minister and the state of public services are really hard to evaluate by themselves. Is the current growth the UK is experiencing a hard-won victory or the absolute minimum that should be expected? Is the NHS world class or failing? Is David Cameron an impressive leader? Even the expert answers to these questions involve a lot of guesswork.

However, in the run up to an election, you don’t have to evaluate things by themselves. Not “how good is the economy?” but “are Labour or Conservative better for the economy?” And – as with hiring programmers on the basis of their programming skills – I suspect that as soon as the economy, political leadership and public services are easier to evaluate, they become more important than immigration, a midterm issue that dominates midterm elections.

It’s a theory, and it’s one that cheers me up when people on the left or right call for the next eleven months to be filled with politicians talking about immigration.


*Immigration is a strange one for New Labour types like me. New Labour’s impetus was the critique that that old left’s priorities were out of touch and its solutions were impractical. Today, being anti-immigration is deeply impractical and it is also quite popular.

**Though this can be mis-read: people are asked to name the most important issues and the implication is that there are important issues to be named. If the economy becomes less of an issue, something must take its place – so weirdly, good GDP numbers and stories about how much people care about immigration will tend to go hand in hand, even if concern about immigration remains unchanged.




Shafir, E. (1993). Choosing versus rejecting: Why some options are both better and worse than others. Memory and Cognition. 21(4), 546-556.

Hsee, C.K., Loewenstein, G.F., Blount, S., & Bazerman, M.H. (1999). Preference Reversals Between Joint and Separate Evaluations of Options: A Review and Theoretical Analysis. Psychological Bulletin, 125(5), 576-590.