Too much of a good thing? Should global capital flows be pumped back up to their boom levels?

I tend to agree with the FT’s leaders – especially their leaders on the world’s macroeconomic imbalances -- more often than naught. But not always. On Friday an FT leader warned about the risk of financial deglobalization:

“Finance is deglobalising out of fear and because of national policies. Neither will be fully undone without political choices that look unlikely, at least for now …. But unless policymakers come up with better global regulation that works we may have to settle for permanently less globalised finance.” (emphasis added)

That didn’t ring true to me. At least not fully. The tone of the leader seemed to long for a return of the pre-crisis world, one where huge quantities of funds flowed across borders, albeit one with better global regulation. Yet just as trade probably rose to a level that could only be supported if US households continued to run up an unsustainable level of debt, cross-border financial flows likely reached levels that could only be sustained if the global financial system remained over-leveraged.

The goal shouldn’t be to return the boom years, but rather to return to a more sustainable level of cross-border flows -- or at least a system without the excesses that contributed to the current crisis. Remember, the rise in cross-border capital flows prior to the crisis was associated with a rise in the amount of leverage in the system, as a host of institutions tried to support bigger balance sheets without increasing their equity. That rise in leverage sustained a lot of cross border flows.

To be concrete:

US financial institutions sponsored offshore special investment vehicles (SIVs) that often borrowed short-term from American investors to buy longer-term US debt. They were offshore largely because they were off balance sheet. If the same activity had been performed on the banks domestic balance sheet – with more short-term wholesale borrowing to support a larger securities book – cross-border flows would fall. But regulators also would have had a lot more information about the build-up of risks in the global system. Taxpayers might not think that is a bad thing.

European institutions seem to have been supporting bigger dollar balance sheets than they could finance entirely in the offshore “euro-dollar” market. Some were borrowing large sums from US money market funds – and then using the proceeds to invest in longer-term US paper. Sometimes securities insured by AIG’s now notorious credit products group, a little trick that allowed the banks to minimize the amount of capital that they had to hold against their dollar book. A bit less of this wouldn’t necessarily be a bad thing. AIG hasn’t worked out so well for the US taxpayer, and the big dollar books of European banks haven’t worked out so well for European taxpayers.

And then, of course, there are the cross-border flows associated with tax avoidance.

The huge rise in transatlantic flows over the past several years produced by the expansion of the cross-border operations of leveraged intermediaries ultimately didn’t produce a more stable financial system -- only a more opaque system. Far less risk from the U.S. housing boom was transferred off the financial sector’s books that most expected. Even some of the risk theoretically shifted to European banks ended up coming back to the US through AIG.

The main reason why cross-border exposure fell in the fourth quarter, incidentally, was a fall in lending among the major economies – not a fall in lending from the major economies to the emerging world. To be sure, lending to the emerging world fell – but the $300 billion fall in exposure to developing countries in the fourth quarter cannot explain the $4800 billion overall fall.

The large two-way global flow between the emerging world and the advanced economies also -- at least in its old guise -- also didn’t necessarily result in a more stable global system.

Take the surge in private money flowing into China. That rise in private inflows was offset by a rise in state outflows. Private money wanted to finance more investment in China, yet China’s government was trying to damp down investment by suppressing domestic credit growth (through the high reserve requirement) and pulling money out of China to invest abroad (through the CIC and other state institutions). Fewer inflows and fewer outflows would result in less total capital flows -- and somewhat slower growth in China’s reserves/ state assets. But that might not be the worst thing in the world, particularly if it was the result of an adjustment in China’s exchange rate that reduced the incentive for speculative inflows into China. A huge rise in private inflows that leads to an offsetting rise in official outflows doesn’t accomplish much, economically speaking -- even it pushes up total global capital flows.

Indeed, smaller aggregate outflows from China wouldn’t necessarily be a bad thing. The large aggregate outflow from China over the past few years hasn’t been driven, generally speaking, by private Chinese demand for the rest of the world’s securities, or even a desire on the part of Chinese savers to diversify their portfolios. Rather Chinese demand for the rest of the world’s financial assets was driven by China’s government, as it bought foreign securities as a byproduct of its effort to hold China’s exchange rate down. Call it subsidized financial globalization: China’s government pulled money out of China and invested it abroad on terms that imply likely financial losses to achieve its policy goals.

Back when money was flowing into China in early 2008 and late 2007, reserve growth and other government outflows were close to 20% of China’s GDP. The gross capital outflow from China then -- almost all a government flow -- was comparable in scale to total gross inflow to the US in the boom years before the August 2007 subprime crisis. Gross inflows to the US peaked in early 2007, incidentally ...

The huge flows of the pre-August 2007 reflected a world where lot of too-big-to-fail financial firms were able to gear up in large part because they were perceived to have a government backstop. These financial flows were in a sense larger than could be supported in a truly unfettered market, as key institutions were operating with a level of leverage and balance sheet opacity that we now know only worked with implicit (and now explicit) government support.

And in other cases, the high level of flows reflected large government-driven flows that were in some sense unnatural. The outflow from China wasn’t a market flow.

A total collapse of global capital flows wouldn’t be a good thing. But returning to the frothy flows of the bubble years -- an era where I think a strong case can be made that a set of distortions pushed aggregate global flows above their natural level -- would not necessarily be a good thing. Not unless cross-border flows rest on a stronger foundation than they do now. The goal should be a happy median, one where a high level of cross-border flows doesn’t reflect a high level of leverage -- -and one where emerging economies don’t just recycle private inflows back into demand for US Treasuries.

The FT leader assumes that absent government pressure more money would be lent across borders, and thus attributes the recent fall in cross-border flows to government intervention. I would start with the opposite presumption: absent government intervention in the third and fourth quarter – the bailout of AIG, big swap lines and the G-7’s commitment to avoid another Lehman – more large global institutions would have collapsed, leading to an even sharper fall in cross-border exposure. A set of financial institutions that were operating globally lacked enough capital to buffer themselves against what should have been a fairly predicable shock;* their near-collapse --not government intervention to prevent their total collapse -- is the main reason why cross-border flows have come to a standstill.

* A large fall in US home prices after a large run up in those prices shouldn’t have been a total surprise.

UPDATE: I edited out a section of the original post that focused on two-way flows to the oil exporters. The post was too long, and the subject is a complex one worth its own post.