What effect will the surge of government bond issuance have on government bond returns?

This is a question that numerous clients and members of the press have asked us so I thought it would be worth writing a brief comment here. 

Focusing on the UK, in yesterday’s budget, chancellor Alistair Darling said that gross gilt issuance will be £220bn this financial year, which is easily a record. There is much speculation as to whether the market is able to digest this much issuance.  If there is a lack of demand, or ‘indigestion’, then prices will have to fall and yields to rise until appetite for gilts returns.  

The chart shows the relationship between gross gilt issuance in each fiscal year since 1991 against the total return from gilts in that period.  There is no relationship.  It’s also a similar story if you measure gilt issuance as a percentage of GDP, or look at net gilt issuance rather than gross gilt issuance.

A lack of correlation is not to say that it doesn’t matter if the supply of government bonds is huge – clearly it does matter.  The law of economics says that if the supply of something increases, then all else being equal, the price will fall. But with regards to government bond issuance, all else does not remain equal.  When governments issue lots of bonds, it generally means that the economy is in trouble.  And if the economy is in trouble, it means that spare capacity is probably being created because unemployment is going up and wages are stagnant (or perhaps even falling).  These things all put downward pressure on inflation.  If inflation falls and interest rates are low or falling, then locking into a high fixed interest rate (at least ‘high’ relative to cash interest rates) is very attractive, and demand for government bonds increases. Yields therefore fall and prices rise.  This is exactly what happened last year in the UK – 2008 was the biggest year for gilt issuance but was still a very good year for gilt returns.

What will cause demand for gilts to rise over the next year to equal or exceed the supply of gilts?  It depends on what happens to inflation and economic growth.  In terms of inflation, Alistair Darling  projects CPI to fall to 1% this year, and RPI to fall to -3% in September before rising to zero next year.  In terms of growth, he expects -3.5% for 2009, +1.25% for 2010 and 3.5% for 2011. 

Slightly ironically, gilt investors should be hoping that the chancellor has overestimated his growth forecasts, even though this will inevitably result in the budget’s numbers not adding up and even more gilts being issued than projected.  Gilt investors should hope Alistair Darling is wrong because if the chancellor is correct about +3.5% growth in 2011, the economy will be booming at its strongest pace since 1999 and you can be pretty confident that government bond yields will be quite a bit higher. 

Much can happen between now and 2011, and his growth projection is certainly possible, but at the moment our view is that it is unlikely that UK growth will be this strong.  If UK economic growth does indeed fall short of his projections, then it’s also likely that inflation will fall short too.  And if your core scenario is that sterling won’t collapse (which would put upwards pressure on inflation), then gilt yields are very capable of going lower.

Finally, as we’ve mentioned previously on this blog, don’t forget what happened to Japan.  There are of course many differences between the UK and Japanese economies, but an important lesson is that large issuance doesn’t mean government bond yields must rise.  The OECD expects Japan’s ratio of public debt to GDP to rise to 197% next year, more than two times as much as for France, Germany and the UK (see chart).  Japanese government debt has tripled since 1996.  And yet today, 10 year Japanese government bonds yield 1.4%, and got as low as 0.4% in 2003.

Discuss Article

  1. Anonymous says:

    I think this is a case of "straw" and "camel’s back", we know the Governments are assuming that if it took anyone 20 years to spot Madoff, then it ought to take at least 50 for them, since they can lie cheat and steal on a global scale. Trouble is, the money supply train is damaging prospects to the extent that even Governments need a wake up call. Remember the single island with a loaf of bread and one pound circulating, which then prints 100 pounds? What is the price of the loaf of bread? Japan has acted in isolation, with blind eyes turned. To use a failed model like Japan’s where the return on assets is zero, because of government intervention and which has been tolerated because it is only defrauding its own citizens, is rather akin to saying it doesn’t matter if you start training all seeing eye dogs to cross in traffic because it got away with it on a quiet country road that is about to be converted to a 6 lane highway.

    Posted on: 24/04/09 | 12:00 am
  2. David T says:

    from april's optimal income commentary:

     "During the month Richard sold 10-year gilt futures, as his view on UK government bonds became more negative on concerns over increased supply"

    doesn't seem to fit that well with the blog.

    Posted on: 20/05/09 | 12:00 am
  3. Chris Ramsden says:

    Thanks for the information on this brilliant blog. This is my first comment but as you can see I am trusting you with part of my pension fund so please take care of it:)

    On the subject matter the BOE currently has an inflation target of 1-3%. this is at a time when currentish indices are RPI (which includes housing) at less than 0% and CPI at greater than 3%. Hence two ways of measuring the same thing span the target range!!!. In addition BOE uses CPI for a target but IL gilts track RPI!! This is like flying a plane with two definitions of horizontal.

    It seems to me that inflation has a number of independent components that should be independently targeted.

    – House price (asset bubble) inflation: Extremely dangerous. Always enjoyed and ignored until the roof caves in.

    – Wage/ price inflation. Watched extremely carefully like generals refighting the 1970s wars. Leads to BOE keeping interest rates too high for too long when price rises occur (even if wages don’t rise) (eg early 2008)

    – Commodity inflation. Not much BOE can do about this. However it leads to above as prices rise without wages.

    – Import price reduction. (eg Chinese goods early 2000s) Compensates for dangerous inflation and leads to BOE holding interest rates too low for too long (early 2000s)

    Wouldn’t it be better to target these different and contrary components independently

    Also, many commentators talk of deflation being always a problem. Well the reduction in fuel prices seems to me an undeniable good thing (since we have now used up most of the oil). I have never known anyone deciding to defer the purchase of significant quantities of petrol (or food) until tomorrow because the price might drop!! Some things have to be bought today.

    I would be interested in your comments. the quality of this site is one of the reasons I invest in your funds.

    Keep up the good work.


    Posted on: 09/06/09 | 12:00 am
  4. Mike Riddell says:

    Defining 'inflation' and deciding what to target is a controversial topic.  If central banks aren't doing either correctly right now, then the implication is that interest rates may be set at the wrong level.

    You're absolutely right about 'good' inflation/deflation and 'bad' inflation/deflation.  If deflation is due to energy and food prices falling then it's good. Likewise, if deflation is due to increases in productivity (eg US economy in late 19th century) then that's very good.  But if deflation is due to consumers reining in spending because they're too indebted, then that's bad because deflation increases the real value of debt and makes things worse.  And if these consumers observe falling prices and then delay spending, then it's very bad because a vicious downward economic spiral can develop. 

    As for targeting different measures of inflation, it's nice in theory but less easy in practice  The overall level of inflation takes account of things like wages, the price of manufacturing and (in this country) food and energy prices so they all contribute to the overall inflation level.  The different components can move in opposite directions, so targeting individual components doesn't really work.  And it makes things even harder if you want to start targeting asset price bubbles (which is very hard since it's difficult to identify a bubble until it's popped).  Even if policy makers could easily identify bubbles, it doesn't necessarily solve the problem of where rates should be.  We had a house price bubble up until H2 2007, so interest rates should probably have been higher than they were. But then wage growth has almost non existent in most of the developed world this decade, which means rates should have been lower.  

    All that said, in practice, central banks do try to balance up the different components.  They typically take a 2-3 year view, so if they thought there was a big risk of a house price crash and a great recession one year in the future, you'd think that they'd not hike rates when perhaps the data at that point in time suggested that they should. You can see evidence of this in the UK – in the first half of 2007,UK interest rates did rise 3 times as the economy was booming, but they didn't continue increasing rates when UK inflation shot above the 3% upper limit for CPI in 2008.  When inflation was at its peak in H2 2008, they were actually cutting rates.

    Posted on: 09/06/09 | 12:00 am

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.