http://www.nzherald.co.nz/immigration/news/article.cfm?c_id=231&objectid=10633293

More seriously, perhaps a better way to phrase the question, Jim, is “why does it take so long for qualified migrants to adjust in New Zealand?”

The research indicates a range of factors including English-language ability, lack of recognition of overseas qualifications and employer discrimination all play a role See: http://www.educationcounts.govt.nz/publications/tertiary_education/55973/1
and:
http://www.educationcounts.govt.nz/publications/tertiary_education/70347/70350

While these factors are all well documented (with varying degrees of evidence), the question that remains as to how could NZ better utilise large number of skilled people who have arrived here in the last 10 years to raise productivity?

Sorry – I don’t have a ready answer to that one.

A lot of our recent growth in working-age people with degrees has come through migration. The research on migration is fairly clear that migrants take anything from 5 to 30 years to attain the same level of income and employment as native born residents – which you could also read ‘to be as productive as’.

So there is another question behind these statistics on the effect of relying on skilled migration to fill skill gaps, particularly on short-run productivity.

And I definitely take your last point about productivity stats – am currently writing a post for my other blog on that very theme . . .

Apples and oranges are still types of fruit, they are just different

The difference between the two gives the minimum contribution to productivity of the increase in skills in NZ. It is the minimum, as some of the contribution is implicitly captured in other measures.

The main point is, these are just different ways of measuring a “labour unit”. We have to define what we want to know before we can decide which measure to use, as the “labour unit” in each case differs. That is what I meant by the apples and oranges thingy.

Ultimately, I would be cautious using productivity stats to account for anything to do with policy. Productivity statistics are notoriously random, so I would try to avoid reading much into them.

Are they really completely “apples and oranges”, Matt? Or is the difference between the labour-volume measure and the composition-adjusted measure the estimated contribution of increased skills to productivity?

In which case, in the table on page 10, increasing skills (or at least quals) account for 0.3 percentage points of productivity growth on average over the 1999–2009 period?

Once we adjust the labour volume index for the fact that skills are higher, it turns out that real labour productivity is lower than we otherwise would have thought. We don’t actually have a counterfactual – we don’t have the case where skills are lower and output adjusts, so we can’t make a conclusion about how skills impact on productivity.

All this measure is doing is explicitly including skills, and since skills are rising it implies that real productivity is lower – it doesn’t mean that skilled labour is less productive.

Note that it says:

“the implication is that composition-adjusted productivity is growing slower than the other productivity measures”

It is because those other productivity measures don’t properly capture skills, and so they don’t properly measure the “labour input” per see.

Comparing the measures is comparing apples with oranges.

Dave, even if labour input is adjusted for quality, it would seem reasonable to expect some higher outputs from higher skilled people.