ADB Thinks Emerging Asia Should Think About Capital Controls

Seems we're getting an outbreak of sanity here, especially after the IMF's change of heart (excerpt of press release):

Emerging Asia Should Be Ready to Act on Potentially Destabilizing Capital Inflows - ADB Report

The report said recent surges in capital inflows have been driven by portfolio equity flows as investors take advantage of widening earnings potential between emerging Asian and mature markets.

The use of capital controls may be appropriate in circumstances where capital inflows are transitory and are adding undue pressure on exchange rates and where effectiveness of macroeconomic policy measures to counter the inflows and the exchange rate movements is uncertain.

"Managing capital flows requires a wide array of policy measures; sound macroeconomic management, a flexible exchange rate regime, a resilient financial system and sometimes the use of temporary and targeted capital controls," Mr. Madhur said.

How and why do capital flows cause problems? There are a number of dimensions to this question, and most of the answers are bad.

First, on an overall basis, there’s little evidence that an open capital account, where portfolio capital is allowed to freely move into or out of a country, has any beneficial effect on economic growth or development (as opposed to capital market growth and development). In fact the opposite effect is more prevalent – liberalisation of the capital account in developing countries has a distressing tendency to turn into a full-fledged currency/financial crisis.

There may be an argument that inflows of foreign capital can lower the cost of capital, and thus increase real investment, but I find this line of thought unpersuasive. As I pointed out once, investment in the stock market is not investment in an economic sense, and does not contribute to a nation’s stock of real wealth – unless shares and bonds are bought at source via the primary market i.e. IPOs or auctions.

More to the point, the real problem with foreign portfolio inflows is twofold – they can just as easily leave as they come in; and they have a destabilising impact on nominal (but not real) valuations. The first problem is well-known, and complicates everything in macro-economic management, from volatility in exchange rates, money supply, and market interest rates, to the increased financial fragility of the financial system through cheap short-term foreign funding recycled through the banking system into long-term illiquid assets such as loans. This is exactly the problem that underlay the vulnerability of Malaysia and other East Asian economies in the run-up to the 1997-98 crisis.

The valuation issue essentially arises from this – nominal asset prices on the capital markets are determined by demand and supply of those assets. There’s nothing here to relate these prices to the underlying intrinsic value of any asset. If an asset was already correctly priced, then incoming money flows would inflate the price of that asset beyond its fundamentals – you have an asset price bubble.

The problem is further complicated by portfolio asset allocation, where foreign fund managers follow certain allocation rules – they generally go for highly liquid stocks and bonds (the better to sell in case of trouble), which in our case would be KLCI component stocks, and MGS or AAA corporate bonds. That means that incoming hot money flows tend to overinflate prices of highly liquid assets, and disproportionately reduce them when they leave. So what you get is much higher volatility in highly visible capital market securities.

The problem is much worse if capital inflows stick around for too long, and get recycled through the banking system. Since capital inflows in the presence of a flexible exchange rate cause an appreciation of the currency, that makes foreign funding cheaper for domestic companies. This is even more apposite since BNM is ahead of the curve in raising official interest rates. Foreign money inflows also have the impact of increasing the money supply and reducing market interest rates, potentially opening the door for inflation.

But the flip side is also true – a sudden outflow of portfolio funds will cause the exchange rate to depreciate, which while it won’t affect your funding cost would certainly affect the principal you have to pay back (not to mention creating a constricted monetary environment just when you need it the least). That in essence was Thailand’s and Indonesia’s bugbear in 1997-98, as their private companies had borrowed in USD but had their income streams in Baht and Rupiah. The downfall of the US dollar pegs in those countries caused a massive private debt crisis, to go along with recession and the loss of international purchasing power.

In short, portfolio capital flows tend to exacerbate changes in the business cycle, which has obvious costs in terms of misallocation of real investment.

So are capital controls the answer?

Only an imperfect one, as in it’s really hard to make them truly effective – where there’s profit to be made, there’s bound to be someone willing to take a risk on both sides of the border (China and its “A” shares are a good example). On the other hand, there is evidence to suggest that capital controls on portfolio capital flows does not harm foreign direct investment at all, so that is a positive.

Second is the issue of whether the capital controls imposed are transitory or permanent – the former can make your policymakers look wishy-washy (low credibility), while the latter would hamper development of the domestic capital markets. You have to pick your poison here.

Third, there’s little precedent for designing effective capital controls in the floating exchange rate era. We have very few case studies of effective implementation of capital controls in the last 40 years, and ironically Malaysia’s two flirtations with capital controls (1993-94, 1998-2005) are among the better known and most studied. Chile in the 1990s is another good example, targeting very specifically short term capital flows. The IMF is now advocating more research into policy design in this area, so we might have some hypotheses to test out soon.