« The Government's Role in Wealth Creation and Redistribution | Main | FOMC Meeting Minutes Leave Room for More Rate Hikes »

Tuesday, February 21, 2006

Debating Dark Matter: Buiter Responds to Hausmann

Here's Willem Buiter's response to Ricardo Hausmann's defense of dark matter:

Martin Wolf's Economist's Forum, by Willem Buiter: Dear Ricardo, dear All,

...I stress two points in my discussion of past and prospective future developments of the US net foreign asset position and net foreign investment income... First, it is not wise to assume that the total risk-adjusted rate of return ... on US foreign assets will in the future be systematically higher than the rate of return on US foreign liabilities. Second, evidence on the historical behaviour of these rates of return since 1980 presented by Hausmann and Sturzenegger ... is deeply suspect.

My note was actually supportive of the H&S Dark Matter hypothesis to the extent that it reported the only robust evidence on the existence of Dark Matter: the stock of US currency held abroad, estimated ... at between US$ 210bn and US$ 525bn at the end of 2004. This is a perpetual zero nominal interest rate loan to the US government by the rest of the world – a nice little earner...

The second category of Dark Matter identified by Hausmann and Sturzenegger are US holdings of emerging market debt (assumed to yield 8 percent for illustrative purposes by H &S) financed with US Treasuries (assumed to yield 5 percent). Identifying the yield differential between emerging market debt and US Treasuries with an expected excess return is plain wrong for two reasons. First, the ex-ante or expected rate of return on emerging market debt ... includes a margin to cover the expected default risk. With reasonable numbers for default risk and default cost, there is actually little room left for a true risk-premium... If default is an event that is both rare and costly when it occurs, the realised returns from holding emerging market debt during a period when no default actually occurs ... will exceed the expected returns on emerging market bonds. The lesson from such ‘Peso problems’ is that past realised rates of return may be poor guides to future expected returns ...

More important, the interest income data recorded in the investment income account are certain to overstate the actual interest income received by US investors holding emerging market debt. This is because of the simple fact, ignored by all other participants in this discussion, that interest is recorded on an accrual basis, not on a cash basis, in the balance of payments accounts. ... Hausmann is plain wrong when he asserts that the accrual problem only causes mismeasurement during the year(s) that the interest is accrued but not paid. At the end of the day, the interest measure in the income account will not be fine. The unpaid accrued interest never gets debited to the investment income account. Instead it gets debited to the capital account. ...

The final category of Dark Matter in the H&S analysis concerns the relative performance of US Direct Investment Abroad (USDIA) and Foreign Direct Investment in the US (FDIUS). The key fact to keep in mind about the market value data on FDI that play such a significant role in this discussion is that these are not data at all but made-up numbers. The equity obtained through FDI is generally not traded and often involves shares of companies that are not even listed. ...

It is clear that the market value of FDIUS is likely to be understated... The reason is that FDI anywhere indeed includes Dark Matter, through the control rights it brings and the often associated transfers of know how, skills and technology. In a reasonably efficient financial market like the US, portfolio equity should therefore, cet. par. have a lower risk-adjusted expected return than foreign direct investment in the US. One can see this for instance, in the superior performance of foreign direct investors in automobile manufacturing in the US, including Nissan, Toyota, Honda, Hyundai and Mercedes, compared to the miserable performance of domestic US automobile producers such as GM and Ford.

Not so in many of the emerging markets and developing countries to which a good chunk of USDIA is directed. It remains true, even in high-risk emerging markets, that FDI has associated with it control rights and technology, knowledge and skill transfers that are absent from portfolio equity investment. The positive excess returns that this conventional FDI dark matter generates ... are, however, likely to be dominated in many of the high-risk emerging markets, by negative excess returns caused by the greater vulnerability of FDI to predatory behaviour by the state or by private sector rivals. Insecure property rights, absence of the rule of law in commercial relations, unpunished illegal or extra-legal predatory behaviour by agents of the state or by private but better-connected competitors are less likely to affect the few companies that have managed to obtain a listing on the domestic stock markets. ... Listed companies in emerging markets are not randomly selected with respect to the factors that make for higher returns. ...

Is it impossible that there is a massive amount of Dark Matter hidden in the US external balance sheet? Everything is possible, but not everything is likely. As regards the contributions by H&S, my bottom line is the following: (1) except for the seigniorage component of Dark Matter, they have not provided us with any hard facts to support their case for massive exports of Dark Matter by the US in the past; (2) as regards future Dark Matter exports, the only sound advice is: don’t count on it. Paraphrasing something Ed McKelvey of Goldman Sachs wrote quite recently: blind faith in the existence of Dark Matter as an excuse for taking a relaxed attitude about the financial deficits of the combined US private and public sectors, would be evidence of a lack of Grey Matter.

    Posted by on Tuesday, February 21, 2006 at 10:33 AM in Economics, International Finance, International Trade | Permalink  TrackBack (0)  Comments (12)


    TrackBack URL for this entry:

    Listed below are links to weblogs that reference Debating Dark Matter: Buiter Responds to Hausmann:


    Feed You can follow this conversation by subscribing to the comment feed for this post.