IMF follows the money in Brazil

Brazil needs massive infrastructure investment

The wrong kind of money. That's Brazil's problem, I'm starting to think, at the start of an IMF conference in Rio about capital inflows, and when and how governments ought to try to control them.

Remember all that talk about "currency wars" at the end of last year? I banged on a lot about it, in the lead up to the G20 Summit in Seoul (for a reminder see my blog from 13 October or my entry two days later).

If you do, you might also remember that the Brazilian Finance Minister, Guido Mantega, was one of the first to use the term, crying foul at rich countries like America exporting their troubles to the emerging markets.

The claim was that, by printing money and keeping interest rates at record lows, the US and other advanced economies were pushing down their currencies against fast-growing countries like Brazil, and sending a wave of cheap liquidity their way which they couldn't safely absorb.

You can debate how much Ben Bernanke has contributed to the flood of capital hitting emerging market shores. But a flood there has certainly been. According to the IMF, net inflows of capital are now at an all-time high for some emerging market economies.

The big institutional investors - as well as more risk-loving types - are piling into these countries, not just because they offer higher interest rates than the likes of the UK, but they seem to offer better long-term growth prospects as well.

Ultimately, this is no bad thing. Indeed, it's what is supposed to happen. It's been one of the odd features of this time in the world's history that developing countries like China have ended up lending developed countries like the US vast amounts of capital.

Economics would say the flows ought to go the other way, with savers in mature economies investing money in emerging markets that offer more exciting opportunities.

The trouble, as I said at the start, is that a lot of these inflows are the "wrong sort of money" - at least for countries with still relatively under-developed financial markets.

Brazil is in dire need of massive infrastructure investment if it's going to maintain the growth it's seen over the past few years - it grew by 7.6% in 2010, the fastest since 1985, after declining just 0.6% as a result of the global recession in 2009.

But there are regular power cuts in the major cities, traffic jams are terrible, and nearly all the airports are operating at or above capacity. And that's before you start thinking about the extra stuff they'll need to host the World Cup in 2014 and the Olympics in 2016.

All told, HSBC reckons that Brazil needs to find an extra 3% of GDP a year over the next four years for these and other investments. It would be great if some of that could come from foreigners (though it must be said, this would come with its problems - Brazilians can be touchy about foreigners getting involved in "key" sectors).

Brazil will host the football World Cup in 2014 and the Olympic Games in 2016

The trouble is that this is not the kind of foreign cash that's coming in. What's striking about this round of capital flows to emerging markets - compared to what happened in the early 1990s and the noughties - is how narrow they are. They consist almost entirely of portfolio flows: foreign investors looking to buy shares and other paper assets, not FDI.

For well-developed markets, the more investors the merrier - it all helps to give local markets breadth and liquidity. But for emerging markets, such flows have traditionally been a mixed blessing, exacerbating local credit booms, and pushing up the value of their currency.

For years, the IMF's response to this was: "tough". Since more integrated financial markets were a good thing, more global capital flows had to be a good thing - all it meant was that emerging economy governments needed to work harder to develop their markets and make sure, for example, their domestic financial systems could manage all this cash. Trying to use capital controls to slow the tide, the IMF said, would do more harm than good.

As I've written before, there's been a lot of "evolution" in the fund's approach in the past decade or so, partly from seeing the consequences of capital inflows run amok, and partly because there were quite a few emerging market economies, like Malaysia and Chile, who had ignored the fund's advice and seemed to have done OK.

Now, the official IMF position is that countries can impose certain kinds of capital controls, in certain circumstances. But there's plenty of heated debate internally about how far this support should go - and whether these controls really work. That's what we're all here in Rio to discuss today.

It sounds like an abstruse topic, but it's actually pretty fundamental to the discussion of what the global economy should look like in the wake of the financial crisis.

Some say the lesson of the past few years is that you can't have a truly free global capital market without recurrent crises - because the free flow of capital inevitably leads to massive imbalances, and massive imbalances inevitably mean big crises when they get unwound (see my blog from 21 December).

According to a chapter in this latest OECD Economic Outlook (in which it saw a risk of stagflation for the global recovery), 60% of the 260-odd recent examples of countries receiving large inflows of capital have ended in a sudden and destabilising way, and one in 10 have ended in either a currency crisis or a banking one (maybe both).

Better banking supervision and risk management - all the reforms we talk about endlessly here in the UK to strengthen the financial system - will be one part of making the world a safer place, in rich countries and emerging ones like Brazil.

That would mean these countries were better able to cope with the money that comes in. But it's a very live issue whether these governments should also be able to stop that money coming in in the first place - and if so, at what cost. I'll let you know how I get on.