Friday, 8 April 2016

Black holes and free lunches

This is a response to a blog post and Twitter chat about the persistence of UK and US current accounts deficits. It's a hot topic: see Frances Coppola. The worst case scenario is generally assumed to be capital flight and devaluation: a return to the summer of 1976 when Callaghan went to the IMF for a loan. This post is an attempt to dig a little deeper, and address the politics and ethics: whose free lunch is it anyway?

The biggest problem for any researcher is the poor quality of the balance sheet data. In theory, if you knew the gross position on the financial and capital account, you could run various stress tests on sustainability: currency and asset price shocks; illiquidity; loan defaults; and so on. I can imagine there are wannabe financial stability super-regulators salivating at the thought of this kind of real-time, country-level portfolio analytics. However, the data are poor and the time series are incomplete.

I know of two major efforts to improve the balance sheet data: the External Wealth of Nations dataset from Lane and Milesi-Ferretti, and the Hidden Wealth of Nations dataset by Gabriel Zucman. Having tried, and failed, to work with the underlying datasets, I consider both these efforts akin to the story of David vs Goliath: heroically correcting the data errors that accumulated in international financial statistics under the laisser-faire stewardship of the IMF and UN.

I'm not very familiar with Zucman but, from what I can see (Table T1) he agrees with Lane and Milesi-Ferretti that the official US external position is, effectively, bankruptcy: assets-liabilities net at around -35% of GDP in 2012. However, both Zucman and Lane/Milesi-Ferrretti argue that this is due to the under-reporting of foreign assets. The most plausible explanations for this are tax avoidance (the purchase of domestic assets by non-residents, who are nonetheless UK residents but moving domestic assets offshore) and capital gains on foreign assets (favourable yields, foreign exchange gains and asset price rises). This somewhat out-of-date graph from Lane and  Milesi-Ferretti (2007, p.232, below) shows the world's aggregate current account deficit disappearing into an offshore 'black hole':

At a first order approximation, Lane and Milesi-Ferretti agree with Zucman's estimate that ~8% of World GDP (or $7.6 trillion) that is being held offshore in tax havens: a 'black hole'.

So far, so good. However, the problem with other parts of the 'dark matter' theory is that it has a poor track record for prediction. Consider Hausmann and Sturzenegger (2006) who, just before the GFC, cast 'doubts on the need for a major adjustment of the dollar or a large rebalancing of the global economy'. I suspect that this poor prediction is partly due to methodology: Hausmann and Sturzenegger inferred the size of the 'assets from their returns... (which) is just like valuing a company by calculating its earnings and multiplying by some price-earnings ratio, or valuing a property based on its rental value ' (p.5-6): akin to driving by only using the rear view mirror.

However, Hausmann and Sturzenegger are also naive in their theoretical framework. They think about 'dark matter' not as the offshore spoils of tax avoidance, but as embedded financial services that the US and UK provide to the rest of the world: 'surprises, risk premia and embedded services [insurance and liquidity]' (page 6). Longeran says something similar in his blog: 'the US has an edge in trading assets and making superior foreign direct assets and selling financial “insurance” is a core US competitive advantage'.

There are well-rehearsed heterodox theories that explain an edge in financial trading, where developed countries exploit their information advantages. If financial markets are rigged ('markets for lemons') then the rest of the world will demand reform: Bretton Woods II might be vetoed by the UK and US, but other players can demand centralised clearing, alternative forms of collateral, breaking up trading cartels; and so on.

What about other embedded services: insurance and liquidity? Does the rest of the world pay a fair price for these? According to Godley and Lavoie (2007), provided UK and US trading partners accept IOUs, they can always issue new IOUs in return for goods and services. There is no need for a Walrasian auctioneer, or equilibrium. If the UK or US issue securities at a higher yield, such as an expensive nuclear power stations or high yield sub-prime/PFI debt, then they only hasten their demise. There is some evidence, at least for the UK, that an obsession with issuing higher yield private debt is doing exactly that, with direct investment income accounting 'for more than 80% of the increase in the (UK current account) deficit since 2011' (Hamroush et al, 2016).

We can also use exchange rate movements to estimate, in part, the historical insurance premium payable on 'safe haven' assets. The following graph takes the known reserves of major exporters (China, Germany and Japan) and estimates foreign exchange losses (the currency composition of reserves is taken from COFER). For China, these unrealised exchange losses approached $1 trillion during the period of USD and GBP weakness after the GFC:

A similar exercise, using Japan's trade surplus as a proxy for Japanese private holdings of US T Bills (not shown) suggests that private Japanese citizens paid out a similar insurance premium. From 1975 to 2013, the YENUSD exchange rate fell from around 300 to around 90. In simple terms, imagine a Japanese exporter in 1979 who had accepted USD in exchange for a Sony Walkman. After 30 years of low returns, they finally decide to convert back to YEN and spend it: their YEN buying power has fallen by about two-thirds.

The foreign exchange losses are consistent with superior trading by the UK and US, but we come full circle: the Walrasian auctioneer is sluggish, unpredictable, greedy and avoiding tax. The insurance premia for safe havens are expensive, and the auctioneer is constantly demanding privatisations to satiate foreign demand and to feed the 'black hole'. The sustainability question is reduced to something simpler: are the insurance premiums too high, and are offshore 'black holes' desirable or sustainable?


Godley, W. and Lavoie, M. (2007). Monetary Economics: an integrated approach to credit, money, income, production and wealth. Palgrave MacMillan. Basingstoke, Hampshire, UK.

Hamroush, S., Luff, M., Banks, A. and Hardie, M. (2016). An analysis of the drivers behind the fall in direct investment earnings and their impact on the UK's current account deficit. Office for National Statistics. Available from: 

Hausmann, R. and Sturzenegger, F. (2006). Global imbalances or bad accounting? The missing dark matter in the wealth of nations ( No. 124). Centre for International Development at Harvard University, Centre for International Development. Cambridge, Mass. Available from:

Lane, P. and Milesi-Ferretti, G. (2007). The external wealth of nations mark II: revised and extended estimates of foreign assets and liabilities, 1970-2004. Journal of International Economics. 73(2007): pp223-250.

No comments:

Post a Comment