Treasure the Treasury Bond

There has obviously been quite a lot of chat about whether there is a bond bubble, but much of this has focused on the supply side of the equation. Evidently, with trillion-dollar deficits in the US and the increasing fiscal costs of bank bail-outs in many Western countries, many are wary of the ability of governments to issue debt to fund these rescues, as witnessed with the dramatic widening in EMU bond spreads and sovereign CDS. But what about the demand side? Firstly, various implementations of the Taylor rule argue for a Fed Funds rate that is significantly negative. With the output gap being so wide there is a clear case that bonds are undervalued, and although extrapolating historical relationships is always risky, bond yields have traditionally not bottomed until the output gap has peaked, and the Fed has not begun tightening monetary policy until the Unemployment rate has peaked (there is certainly a political aspect to this). In such an environment, bonds are likely to be supported.

Secondly, in both the Great Depression and in Japan's Great Recession, the explosion of government deficits did not result in higher bond yields. And, of course, the question is "Why not?". For this, I highly recommend reading "The Holy Grail of Macroeconomics: Japan's Great Recession" (http://astore.amazon.co.uk/breakingviews-21/detail/0470823879) by Nomura's Richard Koo who is arguably the expert on all things to do with Balance Sheet recessions, and Japan's in particular. The basic crux of his argument is that when faced with weak or decimated balance sheets, the private sector ceases to maximise profit, and instead concentrates on paying down debt and repairing its balance sheet. Under such a scenario, private demand for credit collapses and, as a result, domestic credit contracts regardless of the monetary policy that the central bank follows (it does not matter what the cost of funds is if no-one wants to borrow them).

Textbook economics suggests that increases in the monetary base lead to increases in M2 and then to increases in domestic credit via the money multiplier. In the below chart of Japan's experiences, in the late 1980s these three measures grew in tandem until mid-1991 (white vertical line) when policy became too tight and monetary base shrunk and faced with the rising cost of borrowing in the bond market, corporates ramped up their borrowing from banks. When the BoJ finally began to ease, it was too late, and in late 1992 (red vertical line) corporate behaviour changed, and firms began to pay down debt as they realised that their balance sheets had been decimated, leading to a contraction of domestic credit (green line) which continued to shrink until 2005 despite the significant growth in the money supply under QE. Clearly, firms were not interested in borrowing, and thus the very large fiscal stimulae the government enacted throughout the stagnation did not result in inflationary pressure because there was no crowding out of the private sector. This also explains to some extent why QE was not particularly successful in stimulating investment. The corollary to this is that the private sector posted a financial surplus as debt was paid down, resulting in a net inflow of cash into the banking system. In the absense of demand for loans, the banks faced no alternative but to park the money in JGBs - the government was the only entity willing to borrow to invest, and thus was effectively funded by depositors.

And this is what I think is key for the US, UK etc. Rather than being primarily a corporate issue as in Japan (ok, households were also severely affected, but the main problem was corporates), the Western crisis is primarily a household balance sheet issue (in the aftermath of the DotCom bust, firms repaired their balance sheets and generally entered this downturn with reasonable interest coverage from earnings). Faced with negative equity (or the potential to be in negative equity) consumer credit has begun to fall, and with it, consumption has collapsed. The below chart shows the US equivalent of the above Japan chart. As is well-known, the Federal Reserve has ballooned its balance sheet (as illustrated by the jump in the monetary base - red line) and as a result, M2 has begun to accelerate too, but consumer credit (yellow line) has begun to fall as demand for credit from households is falling:The Fed's Senior Loan Officer survey is also showing evidence of this, and the recent credit crunch appears to have been masking this effect:

The credit boom induced commercial banks to reduce their share of holdings of US Treasuries to a record low, but as in past recessions, this fraction tends to increase naturally. However, with households on course to dramatically increase their savings rate and pay down debt, this financial surplus is ever more likely to end up parked in Treasuries. As the below chart suggests, there is no reason why Commercial bank holdings cannot increase to around 20% of outstanding public debt (c.f. China holds ~12%), which provides additional demand of over $1trn.
In conclusion, while there are likely to be occasions when the market gets excited about a turn-up in the data (and eventual recovery), until the output gap peaks, bonds are a buy on dips. The fiscal stimulus is likely to provide such a dip as Q2 GDP is likely to print ~2.5%, and lead the market to price in a false V-shaped recovery, similarly to last spring. Crucially, the above argument does *not* rely on Fed purchases of USTs to create incremental demand (as the market seems to have got itself into a tizz about) - the demand will come from commercial banks themselves.