There is, however, another explanation. Index-linked yields are low because labour productivity growth has slumped.
There has for years been a close correlation between productivity growth and index-linked yields. In the early 90s when GDP per worker-hour was growing quickly - by over 3 per cent a year - index-linked gilts yielded almost 4 per cent. But as productivity growth has trended downwards, so have yields.
There's a simple reason for this. Low productivity growth means that companies cannot expand without incurring the cost of taking on more workers. This implies that there's a danger either of inflation, as companies raise prices to cover rising costs, or of a squeeze on profit margins, or of low growth as companies decide against expansion.
In this environment, index-linked gilts are attractive. Unlike conventional gilts, they protect us from inflation. And unlike equities they protect us from disappointing real growth or a profit squeeze. Low productivity growth thus creates high demand for index-linked gilts, causing yields to be low.
If we'd had index-linked gilts in the 1970s - they were not issued until 1981 - there would have been massive demand for them, because stagflation (weak growth and stubborn inflation) is the perfect environment for them.
We can put this another way. When productivity growth is high, the prospects for inflation-free growth are good, and so equities are attractive. Yields on index-linked gilts must therefore be high in order to tempt investors to continue holding them when equities are so attractive. As equities have lost this attraction, however, so investors have no longer needed high yields to tempt them to hold index-linked gilts.
You might wonder how we can reconcile this account of low yields with the conventional story I began with. Simple. The savings glut that has driven yields down is the counterpart of a dearth of investment opportunities; excess savings is another way of saying 'deficient investment'. And the investment dearth and productivity slowdown are two facets of the same problem. The slowdown in the rate of technical progress that has contributed to the productivity slowdown has also reduced monetisable investment opportunities.
If technical progress and productivity growth had been faster in the 00s, companies - in the west generally, not just the UK - would have invested more in real assets. This would have diverted savings away from government bonds, the housing market and mortgage derivatives. Bond yields would therefore have been higher and we might not have had the banking crisis that has driven gilt yields down so far.
In this sense, the productivity slowdown, and the slowdown in technical progress that lies behind it, is the central fact of the investment climate in recent years. Productivity statistics might not drive markets on a day-to-day basis. But they set the climate, if not the weather.
All this poses the question: what would happen if or when productivity growth picks up?
One possibility is that this would happen if companies that have been hoarding labour in the hope of an upturn in demand finally decide to throw in the towel and shed staff, or if 'zombie companies' - those inefficient ones struggling along thanks to banks forbearance - are destroyed. In this environment, unemployment would rise and conventional gilts would do well both in absolute terms - as investors seek a safe haven against a weak economy - and relative to index-linked gilts as inflation expectations come down.
The more benign possibility is that a recovery in demand causes the pro-cyclical pick up in productivity growth that we usually see in economic upswings. This would cause index-linked yields to rise simply as better economic conditions reduce investors' risk aversion, causing them to shift out of index-linked gilts and into shares.
In fact, these two scenarios aren't mutually exclusive. Some companies might benefit from increased demand while others shed staff or go bust, causing unemployment to rise even as the economy picks up. This would favour equities and conventional gilts over index-linked.
However, while these are both likely to happen sometime, there seems little chance of them happening quickly. In this sense, index-linked yields could stay negative for a few more months.
