Prof. Dani Rodik re-invents Selective Employment Tax?

If you’re looking for daft taxes, the so called “Selective Employment Tax” implemented in the UK in 1966 takes some beating. This tax was “selective” in that it was biased in favour of manufacturing and against services.

The rationale was always obscure. Allegedly the UK Labour government of the day was worried about well qualified school and university leavers rejecting manufacturing jobs and taking service sector jobs: like those well paid “service sector jobs” that Labour governments themselves create in ever expanding numbers.

There was also the argument that more manufactured output is exported than is the case with services. Plus there was the argument that output per head expands faster in the long run in manufacturing than in the services sector. Ergo the more manufacturing you have, the faster is economic growth.

The latter argument has re-surfaced in an article on the “Vox” site entitled “Making room for China in the world economy” by Prof. Dani Rodrik. He advocates some sort of bias in favour of manufacturing for China: an over valued currency or some sort of subsidy for manufacturing.

To illustrate the flaw in the above “output per hour” argument, let’s start with a closed economy and consider computing. Output per head in this industry (in terms of computing power manufactured per worker-hour) has expanded by about a hundred fold in the last twenty years. But does this mean that it makes sense for everyone to spend half their income on personal computers?

The latter conclusion is to go back to pre-Say economics. Jean-Baptiste Say’s big insight was that the value of products is determined by what the customer wants to pay for them, not by the cost of production (though doubtless plenty of ancient Egyptians tumbled to this one).

I.e. at the margin, worker-hours should be devoted to producing whatever the consumer pays the most for.

Turning now to open economies, the first problem is that the pro-manufacturing argument applies not just to China, but to half the rest of the world. Now if half (or the whole) of the rest of the world have pro-manufacturing subsidies, the subsidies all cancel out! At least the above mentioned “export” argument is severely dented.

The next flaw in the pro-manufacturing argument is the fact that it measures GDP on the basis of how much is produced as measured in terms of the producing country’s own currency. The big weakness here is that if the currency is undervalued, those doing the production are paid less for their efforts than if the currency had a realistic value on world markets.

To illustrate, revalue the Yuan, and ten million Chinese farmers will be able to purchase the TV set or the PC they previously could not afford. This is because the revaluation will cut exports by Chinese electronics goods manufacturers, thus the latter will look for alternative customers. Plus the revaluation makes the Yuan worth more in terms of dollars, thus these manufacturers are likely to find Chinese farmers’ Yuans an acceptable substitute for US households’ dollars.

Rodik's argument is a good illustration of a common mistake in economics: targeting what is sometimes called an “intermediate objective”. The fundamental economic objective is maximising real wages per hour of work (within environmental constraints). The above argument targets PRODUCTION. Production is not the fundamental objective: the fundamental objective is CONSUMPTION.

As Dani Rodrik points out, his ideas fall foul of World Trade Organisation rules. My answer is that there is method in the WTO’s madness – though doubtless large numbers of unemployables screaming students might beg to differ.