Comments on Dooley, Garber and Folkerts-Landau: part 1, China

As usual, Michael Dooley, Peter Garber and David Folkerts-Landau are provocative.

And in a lot areas, the arguments made by those - like Dooley, Garber and Folkerts-Landau -- who argue that the Bretton Woods 2 system will last for another eight years, if not another twenty and the arguments made by those like Eichengreen, Roubini and Setser who argue that the Bretton Woods 2 system is likely to be on its last legs converge.  

I suspect we all would agree that the buildup of reserves by foreign central banks, mostly central banks in Asia in 2003 and 2004 and now a combination of China and the central banks of the oil exporters, plays a big role in the global economy.  I don't think Dooley, Garber and Folkerts-Landau would argue with recent work by the Banque de France (see p.2 of the link) that argues intervention by foreign central banks knocked between 125 bp off long-term Treasury yields in the first part of 2004, and 115 bp in the second half of 2004 --  far more than the Federal Reserve estimates.

But I do not see how Dooley, Garber and Folkerts-Landau can continue to push their offshore intermediation of Chinese savings thesis - what they call the total return swap.    The data out of China just does not fit the "offshore intermediate" thesis.   It worked from say 98 to 2000, but not from 2001 on.

(More follows, with another CR style graph)

I also do not see why smart market analysts continue to refer back to the "offshore intermediation of Chinese savings" part of the Dooley, Garber and Folkerts-Landau argument.  To quote Paul McCulley:

China itself is not very good-in fact, they are very bad-at intermediating their very high savings rate. China has always been a very high savings country. Traditionally, they have intermediated those savings—very poorly—through a state-owned banking system. In many respects, what they are doing now is outsourcing the intermediation of their savings.

The offshore intermediation argument more or less goes like this:

China's banking system is inefficient, and cannot effectively invest China's domestic savings.

Chinese citizens invest their spare savings abroad in the international banking system.

The world's big banks lend China's savings surplus back to foreign firms who invest in China.

And as a result of "outsourcing" financial intermediation, China gets a high quality capital stock.

There is a twist about China needing to run a trade surplus to provide collateral to match the increase in value of past FDI in China, but let's set that part of the argument aside.

Basically, Dooley et al postulate a world where Chinese savings flows abroad - so, leaving FDI aside, there is a net outflow of capital from China.  

That was true from 1998 to 2000, when, in the aftermath of the Asian crisis, there was real concern that China would devalue its currency.   And interest rates on dollar assets were kind of high back then too.  

My problem: It has not been true since 2000.

In 2001, Chinese capital stopped flowing abroad.  And in 2003, existing Chinese savings abroad started coming home in a big way.

The Dooley-Garber-Folkerts-Landau thesis is consistent with the data in the first two year of this graphs (data comes from Prasad and Wei, with Setser updates) but not consistent with the data in 2001 or 2002 - and in no way matches the data for 2003, 2004 or 2005 (2005 obviously is forecast based on data from the first half of the year).    Yes, FDI continues to flow into China.  But these FDI inflows are not  financed in some broad sense by other capital flowing out of China.    Look at the "white" portion of the graph, the non-FDI capital account, from 2003 on.  It has the wrong sign for the Dooley, Garber and Folkerts-Landau argument.

Dooley et al talk of China' moribund financial system.   Yet over the past three years, China's moribund banks have increased their lending by something like $900 billion over the past three years.  That is about 50% of China's GDP.    Not exactly moribund in my books.   I agree - no shock - with Goldstein and Lardy.  The facts in China don't fit the Dooley, Garber, and Folkerts-Landau thesis.  Most capital investment in China is financed domestically, by the banks and by the retained earnings of Chinese firms, not by foreign direct investment.

I can see why folks in the market would back the Dooley, Garber and Folkerts-Landau argument that foreign central banks will continue to accumulate $500 b plus in reserves for far longer than Nouriel and I argue.  That part of their argument is far easier to defend, and certainly is consistent with China's current reserve accumulation.  But so long as hot money is flowing into China (and into Chinese banks), not out of China and out of Chinese banks, the offshore intermediation thesis needs to be retired ...