In our Devalue or Else debate John Mills has been arguing for a concerted policy to lower sterling’s exchange rate in order to achieve a number of goals for the British economy, including now, in his latest article, the reduction of inequality that many agree is desirable:

- it will help rebalance the economy away from services and towards manufacturing

- it will create “good jobs” in manufacturing and so help reduce inequality

- it will reduce unemployment and so will increase the bargaining power of labour, so further reducing inequality.

He has developed his case in his new book, Exchange Rate Alignements. But some of the putative links between the exchange rate and the distribution of economic output are not obviously as benign as he suggests. John Mills explains that the City has always liked a strong pound and low inflation, and its political clout means that this is what we have got. The self-serving reasons for that preference is not entirely clear but might go something like this:

UK financial institutions still need to hold sterling dominated portfolios - both for regulatory reasons (although there are fewer and fewer constraints on what currencies pension funds and insurance companies must hold) and because many of their final customers ask for sterling liquidity

- hedging sterling is a cost - especially if inflation and exchange rates are volatile

- depreciations reduce the value of sterling assets in world markets, and so reduce the firepower of UK institutions in world financial markets - a UK bank that needs to satisfy various internationally set liquidity constraints will find it harder to meet these if their sterling assets depreciate

However, in a world of pervasive currency hedging, it is not obvious that this kind of preference should be particularly strong. And as John Mills notes, financial services themselves have not been very sensitive to the price charged - which suggests that the financial sector might actually not - at least  from a perspective of pure self-interest - find a strong pound such a necessity. The sorts of financial services that the City specialises in are internationally traded and their price is set in world markets. Whatever happens to the pound, the buying power in the hands of the City operators is unlikely to change very much. If the pound depreciates, the hot-shots will soon negotiate their pay packets back up to what world financial markets will bear.

All this suggests to me that a lower pound might not be so painful to the City, and might in fact increase inequality - at least it would increase the pound value of the pay packets at the very top, those of the foot-loose who’ll negotiate making credible threats to decamp to Zurich, Dubai or Singapore.

Now that might not matter much - the detached global elite is exactly the one that has made itself almost untouchable by any single national policy. They’re at home in resorts and international airport lounges; in brand and boutique hotels; their needs are catered for in every corner of the world; cities fall over themselves to be attractive to this high-spending, fine-living class. Here is a group that has grown in the opportunities offered by a global economy characterised by internationally mobile capital and nationally bounded taxation and regulation. Exchange rate policy isn’t going to touch this group much.

What will touch it - or a particularly egregious subset of it - is for us to sort out financial regulation and to make sure that the sector is not a giant magnet for un-acknowledged subsidies, especially in the form of the lender-of-last resort guarantees that have underwritten the casino behaviour of the banks and quasi-banks.

So the beneficial distributional consequences of John’s policy are not likely to come from reducing the take at the very top. But they could come from increasing the take at the bottom or in the middle. John, in his response to Robert Skidelski, argues that:

“If we then need to rebalance the economy with new manufacturing processes, I think we would be far better to concentrate on low tech. Entry is much easier and quicker. The problem with low tech, however, is that ability to compete on world markets depends entirely on the cost base and the exchange rate.”

I take this to mean that John would like to see an exchange rate such that low tech manufacturing - the sort currently taking place in China - become competitive in the UK. John sees this as the way to rapidly create good manufacturing jobs in the UK. Again, the distributional consequences of this policy need to be teased out: there is no reason to believe that the purchasing power in the hands of those employed in the UK in these jobs would be hugely different from that available from this sort of activity anywhere in the world - under free trade in goods, the returns to similar factors of production are likely to be similar. They will differ because of local factors, negotiating strength of organised labour, regulation, and such. But they will not differ by orders of magnitude. Average manufacturing labour costs in China are really very low - somewhere between $1 and $2.50 per hour in 2008. The same article estimates that EU costs are 30 times higher in manufacturing than in China. Of course, these are not like-for-like hours being measured. However, John recognises that the devaluation needed to be competitive in low tech might well be very substantial.

So if those jobs - same production plant, same technology, same productivity - were to be relocated to the UK, one would not expect the labour component of value added to be massively higher. Some transport costs would be saved, and stronger labour negotiation and labour protection might increase the labour share somewhat. But this would still imply pretty low wages for those employed in these activities. Hence Robert Skidelski’s emphasis, in his response to John, on making sure that high value added jobs are there to be performed by high productivity workers. Without higher productivity than the low-tech jobs being currently done in China, the wages of those employed in these activities will be worryingly low from a distributional point of view.

Those who are already paying their way in world markets will be able to insulate themselves from the wage-impacts of a devaluation. Indeed, they will feel better off to the extent that their non-traded input costs and purchases will have fallen in price. Those employed in the newly competitive low-tech manufacturing will notice that each pound does not stretch very far - less gas and electricity, a less good car, even if the price of a hair-cut has not increased much.

Of course, it may be that there is no choice about all this. In the end, if the UK is not paying its way internationally - and John points out that we are only settling our net import bills by continued borrowing and running down of assets - then we’re going to have to buy fewer foreign-made goods, and we’ll collectively feel the poorer for it. It may well be that Robert Skidelski’s hoped-for industrial policy that creates high quality jobs for highly productive workers is a pipe-dream, nostalgia for a bye-gone era of easy growth. John may well be right that this is a bullet we must bite. If it is, he is probably right that there is no point in delaying. But I don’t think he’s right that the social consequences will be all - or even mainly - nice.