Sunday, 9 February 2014

The Grand Old Duke of Rio

The emerging market sell-off paused for much of last week but the debate about what it all 'means' rages on.  So to carry on from last weekend's themes here are a few points and a picture:

Firstly, a re-cap: really low rates (and QE) in the US (and Europe) caused a huge flow of money into almost any asset with any yield both in 2004-2008 and 2009-2002. Secondly, many (not all) central bankers allowed the effects of this to be felt in 2009-2012 through currency appreciation, a major difference with the 1990s. Some of them ended up with very expensive currencies. The US ended up with a very cheap one. Thirdly, many (but definitely not all) EM countries were consequently able to avoid the explosion of debt, notably foreign currency debt, that they suffered from in the 1990s. In aggregate, foreign currency debt is a smaller share of EM exports today that it was a decade ago and the interest that is paid on it totals something like 2 1/2% of export earnings. Fourthly, it isn't the US' 'fault' that the Fed sets a domestic level of interest rates that works for the US but doesn't work for the rest of the world. And finally, I think it would be very useful if we stopped talking EM and DM, and at the very least took China out of EM and started calling it the world's second biggest economy with the world's biggest domestic credit problem. It makes a mess of 'EM-wide' data if it isn't considered separately and if China's debt bubble does burst, I for one won't describe the outcome as an 'EM crisis'  

Anyway, here's a chart of real effective exchange rates, some EM and some DM, back to 2005. I drew it to keep my mind occupied for the last hour of a truly awful game of football yesterday and I'll do some 'work' on this theme in the next few days.  The data are from the BIS, and I've updated them as best as I could in the time I had. With any index-based look at FX, the starting point is arbitrary and therefore prone to skewing conclusions if you're not careful, but I chose 10 years ago because that was far enough after the 1990s EM crisis for recovery to be underway, and far enough after the Euro's launch for EUR/USD to have both collapsed and recovered. So for better or for worse, this chart shows how real exchange rates have moved since the global economy was in a relatively balanced state, in the early stages of the great financial bubble.

A few things are interesting. The first is that towards the end of 2012, the three cheapest currencies on this arbitrary list and relative to my arbitrary starting point, are the dollar, euro and pound. The second is that three currencies saw truly staggering real appreciation in 2004-2008 and then again after 2009:  The Chinese Yuan has appreciated by 40% in real terms over a decade. The Brazilian real had at one point almost doubled in value in real terms and even now is still 50% more expensive than it was in 2004. Go to the World Cup, and experience what that means for yourself. The Rouble too, is 40% more expensive in real terms than it was. If you are in Sochi, you probably realise what that means. But at the other end of the spectrum, whatever else is going wrong in South Africa, the rand has replaced the pound as the biggest FX faller in this list over the last decade, the Turkish Lira is correcting fast and the whole EMFX boom, in real terms anyway, rather passed the Mexico Peso by.

Perhaps it isn't surprising given the trend above, that the current account balance of the developed economies as a whole, should have improved by over USD 400bn per annum over this period. The developed economies now have a combined current account surplus. And by contrast, the surplus of the emerging and developing economies has, of course, shrunk. Indeed, in all probability some time this year the current account balance of 'EM' will be in deficit if we exclude China.

The shift in the balance of payments reflects in part the real appreciation of EM currencies and the real depreciation of the two most important developed economy currencies. But that still doesn't automatically mean that EM disaster in upon us. In 2007, just before everything went wrong, the EM world had a $600bn current account surplus, but an even bigger inflow of private capital, almost USD 700bn. The other side of that coin was a USD 1.2trn increase in EM central banks' currency reserves. Madness! We all thought everything would change after 2008 but in 2010-2012, the total private sector inflow into EM was $1.5trn, and FX reserves grew by another $2trn to mop as much of that up as possible.

So much investor capital has flowed into EM overall that the global financial system is inherently unstable and prone to the kind of volatility we are seeing. Really, am I supposed to be shocked by weekly flow data that show money still coming out of EM funds?

More facts: External debt of EM economies has risen from around $3trn in 2005 to about $7trn now but as  a share of exports, that total is actually down slightly from 80% to something a little above 75% today. It's above 100% GDP in Latin America and the CIS while in Central and Eastern Europe it's north of 150% of exports. It's around 50% of exports in Asia.

Finally, the big hornet's next is the growth of domestic debt, and this is a global rather than an EM problem. There is much debate about whether deflation is here, whether it's bad and whether there are good and bad kinds of deflation. The kind of deflation that will cause us all trouble will be the kind that results in nominal growth rates being too low to prevent debt/GDP ratios heading inexorably upwards. That's the biggest challenge facing Japan, Europe and as Chinese growth slows while debt grows very quickly, it will be more than a headache for the world's second biggest economy.

Saturday, 1 February 2014

Unravelling the global effects of mad money

Since some of my 'work' observations on currencies have been doing the rounds in the press over the last few days, I thought I'd lay out some broad thoughts on recent weakness in emerging market (and other) currencies. There's a danger in broad generalisation in a quick sweep through thirty years of the after-effects of very low US interest rates, but this is a blog not a research paper so here goes....

In 1993, when the Federal Reserve maintained interest rates at what was then perceived to be an astonishingly low level of 3% as the US economy recovered from the aftermath of the S&L crisis and the recession two years earlier, the seeds were sown for the 1990s Asian crisis.

When the central bank of the world's biggest economy decides that its domestic interests are best served by extraordinary monetary policy settings, policy-makers all over the world have to decide how to react. In 1993, by and large, Asian policy-makers' response was to continue to gradually liberalise the flow of money into and out of their economies, to maintain formal or informal pegs between their currencies and the US dollar and to let the good times roll. In countries like Thailand, the decision not to allow currencies to appreciate pretty much automatically resulted in US interest rates being imported into economies which were growing far too fast for 3% rates. What followed was a surge in both domestic credit growth and in foreign currency borrowing - if your central bank sets rates at 6%, the Fed is at 3% and you have a fixed exchange rate between your currency and the US dollar, why not just borrow cheap dollars?

The result through the early part of the 1990s was that a lot of Asian countries maintained competitive exchange rates and saw very rapid growth in credit. That was fine for a while but as US interest rates rose, it became problematic. It became clear to everyone that what looked like a credit 'boom' was a bubble and that too many golf courses, casinos and beach resorts had been built without regard for whether there was income enough to service the debts to pay for them - debts whose services costs were going up.

As growth slowed, bad debts ballooned and capital left their countries, Asian central banks found that they did not have enough currency reserves to stop the pressure on their (pegged) exchange rates, and were forced to choose between letting them fall (making the foreign currency debt problem worse), raising interest rates (causing an even greater economic downturn) or imposing capital controls. Different countries chose different paths but the Asian crisis followed and the next few years were spent trying to learn the lessons from the disaster.

The lessons which were learnt served Asia well in 2001-2012. As the Federal Reserve once again embarked on a policy of exceptionally easy monetary policy after 2000, Asian central banks were much more alert to the danger of poor lending practises and to the danger of foreign currency borrowing. They set about building vast war-chests of currency reserves to help defend themselves against future runs on their currencies. And they started to allow their currencies to float somewhat more freely, which mostly meant that they allowed them to appreciate against a US dollar anchored by low interest rates.

Because they avoided the worst excesses of credit growth in this period, Asian economies weathered the 2008 crisis much better than Europe and the US. But what came out of that period was even lower interest rates (3% Fed Funds in the 1990s become 1% Fed Funds in the 2000s and then zero Fed Funds in the 2010s) with the added twist of Quantitative Easing.

The challenge remained the same and was no longer in any way an 'Asian' problem or even one just for 'emerging' economies. The central bank of the world's biggest economy set interest rates at zero when the emerging market economies as a whole were enjoying real GDP growth of 7 1/2% in 2010. They faced the choice of allowing their currencies to appreciate sharply, of allowing domestic asset prices to rise and credit to grow because rates were too low, or of imposing capital controls. No wonder many complained they were in a currency war.

Different countries faced different domestic issues and reacted to the global environment in different ways. But if there is a common thread it is that most allowed their currencies, in nominal and even more in real terms, to appreciate more than they had in the early 1990s.  In 1993, a young backpacker wandering the world would generally have found that 'emerging economies' were cheap places to visit. By 2011, that was no longer the case. Again, it was not just an 'emerging market' story. In 1995, British people could go and enjoy cheap prices for winter sun in Australia, while young Australians came to London to earn better pay than they could get at home and see the world at the same time. Now, Australia's youth stays at home to serve beer to locals who enjoy watching their cricket team thrash England's.

When the notion of the Federal Reserve 'tapering' its bond purchases was first mooted last May, there was a temptation to ask whether that was important. Buying bonds more slowly is only a marginal shift and interest rates are still at zero, after all. However, a turn in the monetary policy cycle was signalled and was enough to trigger a violent response in asset markets. US bond yields rose and the currencies which had been allowed to appreciate in reponse to super-easy US monetary policy, started to fall back, chaotically in many cases.

The most positive way to look at the events of the last 9 months is that most of the central banks whose currencies are falling, probably welcome at least part of the adjustment. They didn't choose to have overvalued exchange rates, after all. What they chose was to allow the currency to take the strain of offsetting US monetary policy, which in turn allowed them to maintain some degree of control over domestic credit conditions. If I compare the modest sell-off we have seen in emerging market bonds, for example, with the larger moves we have seen in currencies, the conclusion is that the currency is doing its job of acting as a shock-absorber and this is a currency adjustment (so far) and not a crisis. In Australia, whose currency is now 20% cheaper than it was at its peak, there is no sense of crisis at all. In Argentina whose currency has fallen almost 20% in January, it's definitely a crisis.

Whether it is right to conclude that there is 'no crisis' is a matter of judgement. But I would draw three very broad conclusions:

Firstly, the willingness to allow currencies to act as a shock-absorber helps prevent the extremes of domestic credit imbalances which plagued Asia in the 1990s. That is a very good thing indeed in those countries which did allow their currencies to float more freely and did manage their domestic financial system more closely.

Secondly, given that we are still in the very early stages of the US policy nomalisation and since there are still an awful lot of countries whose currencies are far more expensive in real terms today than they were 10 years ago, it would be rash to assume that the adjustment is mostly complete.

And thirdly, since Fed policy was even more accommodative in this cycle than in the early 2000s or the 1990s, it isn't surprising that the scale of the capital flight from G3 economies into the rest of the world was much bigger this time and the accompanying asset inflation was also much greater. And so, the turmoil that can be caused by those flows being reversed, can also be greater than it was in the second half of the 1990s. Which is to say that even while I don't think this IS a crisis at this point, the risk of it becoming one is clear. Particularly in those countries where control of domestic asset inflation and credit creation was weakest.