Make it so

By Buttonwood

DESPITE the debate about the fiscal problems of America (or indeed Britain) and a sell-off in the last few days, government bond yields are at very low levels. For much of history, governments would have been delighted to be able to borrow for 10 years at 1.8%, as the US Treasury can today.

One way of interpreting those low yields is in terms of the demand for safe assets, as an intriguing paper from the Bank for International Settlements discusses. It is a rich theme. One aspect is the way that Asian central banks built up their foreign exchange reserves after the crisis of the late 1990s, driving down Treasury bond yields. Such low yields fuelled the demand for alternative forms of "safe" assets that were artificially created by the finance sector in the form of the structure products linked to subprime mortgages. The realisation that these assets were not safe at all sparked the debt crisis, and caused a surge in demand for government bonds as the only true safe asset. The process was described in a previous paper by one of the BIS authors, and discussed in a Free Exchange column last year.

The useful justification for fiscal deficits at a time of recession is the Keynesian one of supporting demand. But in the current crisis, one could argue that governments were simply providing a service - supplying a sufficient quantity of safe assets to meet the demands of investors.

But if investors were gagging for government bonds, why did central banks need to intervene in the form of quantitative easing? The BIS paper answers this problem by referring to the natural fears of investors about the long-term solvency of governments. As we have seen in Greece and Portugal, a sharp rise in government bond yields not only hurts the rest of the economy, it also has a self-fulfilling negative effect on the outlook for government finances. When the central bank steps in, the BIS argues that

the monetary backstop stabilizes the dynamics of government debt by preventing negative shocks from raising the interest rate and the ensuing feedback loop between high interest rates, rising debt and pessimistic expectations. Given the terms of the trade-off, it is not clear a priori that it is preferable to minimise the risk of inflation rather than that of default.

It is an elegant defence of QE and the paper also musters an explanation for the difference between ECB and Fed policy. As cynics like your blogger have suggested, the BIS admits that

one can never completely exclude that (a policy which) starts as lending in the last resort turns out to be the first step in the monetisation of an insolvent government.

But it is one thing for the authorities within a single country to take this gamble - one might tolerate quite a lot of inflation in order to avoid economic collapse. Within the euro-zone, however, you are essentially asking the citizens of one country to risk inflation in order to help out another - a much less attractive bet.

All this leads to a discussion of the changing role of the central bank over the ages. The Bank of England was set up in 1694 as a way for King William III to obtain finance for his war against France; in return for a loan to the government, the Bank was given various privileges, including the right of note issuance. So there was a kind of circularity built into the system from the start. What made the Bank powerful? The privileges granted by the government. And what kept the government solvent? The loan provided by the Bank.

In the 19th century, central banks gained two new functions; the role of lender of last resort to the financial system and the guarantor of the currency. These dual roles have always had the potential for conflict and still do today. Under the gold standard, the financial sector might need more support than the central bank could deliver, consistent with the need to keep a consistent ratio between gold and its reserves. (The answer, as in 1890 when Barings collapsed was to borrow money from friendly foreign central banks.)

As the monetary system evolved, the currency guarantor function of the central banks changed with it. The Bank no longer had to worry about having enough gold but it did have to worry about having enough foreign exchange reserves to safeguard the currency against speculative attacks, something the Bank of England failed to do in 1992. The introduction of inflation targeting turned the Bank from the guarantor of the external value of the currency to the guarantor of its internal value. And, of course, some central banks (including the Fed) target economic growth as well as inflation.

It is a lot for an unelected body of experts to do and, in my view, there has not been enough debate about the power of central banks within a democracy. it is not easy for politicians to monitor central banks and the ultimate sanction - dismissing the man in charge - carries such risk of financial and economic damage that few governments would dare to use it. At the other end of the spectrum, it is clear that a government that can persuade a central bank to buy its debt has an enormously powerful weapon; offering services to voters without demanding the taxes to pay for them (in the short term, at least). This power can place a political opposition at a severe disadvantage.

Coming back to the idea of safe assets, how would we define the term? The best description is an asset that protects the purchasing power of investors. Does that apply to government bonds? History suggests not. In nominal terms, governments have defaulted many times, as Reinhart and Rogoff demonstrated at length. As Norman Davies reminded us in his recent book Vanished Kingdoms, history is also littered with countries that ceased to exist (in the last 25 years, Czechoslovakia, the Soviet Union and Yugoslavia have disappeared; the UK might not exist, in its current form, by 2015).

In real terms, an investor who bought Treasury bonds in 1947 had seen their purchasing power fall 90% by 1981. Even if one buys inflation-linked bonds, there is always the risk that the government redefines inflation.

In practice, safe assets are defined in a more technical sense, as those assets that institutions are allowed to treat as safe for accounting or regulatory purposes. Governments can act like Captain Jean-Luc Picard shouting "Make it so" and decree that their own debt be treated as safe. This is, arguably, an even greater power than the ability to tax which is limited by the exigencies of electoral politics and the realities of mobile international capital and labour. But it creates a kind of "government-central bank-commercial bank complex" in which money flows between the various parties, rather like the Escher print in which water flows downhill but ends up back where it started.

Can this system be sustained? The BIS paper sums up the issue by writing that

The authorities should commit themselves to a clear defintion of safe assets and back it with a policy regime that makes those assets credibly safe.

There's the rub. As the authors remark, this looks like a repeat of the Triffin paradox of the 1960s in which the world financial system needed dollars to function but when more dollars were supplied, investors became less convinced of their value. The paradox led to the collapse of the Bretton Woods system. Now the world has an enormous appetite for government bonds, but the more bonds are created, the less safe they appear to be.