I attended an interesting lecture last year by Carmen Reinhart (hosted by RBC) where she predicted an era of negative real returns for investors in developed market sovereign bonds. This was due to what she termed ‘financial repression’, as the authorities in richer countries struggle with the huge debt burden. Financial repression is loosely taken to be things such as greater financial regulation, capital controls designed to limit capital flows, restrictions on competition in the financial sector, credit restrictions, higher bank reserve requirements, enforced liquidity ratios (so banks must hold a minimum of their reserves in government bonds), pension fund legislation forcing pensions to own more domestic sovereign debt, or caps on deposit rates. The consequence of financial repression is that nominal sovereign bond yields are artificially depressed for a prolonged period of time, and domestic investors therefore experience low or negative real returns.
Reinhart and Sbrancia have since elaborated on these thoughts (see here), where they argue that financial repression is a subtle form of debt restructuring and is a policy that had significant success in reducing government debt burdens post the Second World War. Far from being a modern phenomenon or one confined solely to emerging markets, they find that real interest rates in advanced economies were negative about half the time between 1945 and 1980, and ‘for the United States and the United Kingdom, the annual liquidation of debt via negative real interest rates amounted to 3 to 4 percent of GDP on average per year. Such annual deficit reduction quickly accumulates (even without any compounding) to a 30-40 percent of GDP debt reduction in the course of a decade.’
Many investors and commentators point to negative real yields today as a reason why government bonds are unattractive (see here for a sensationalist write-up on negative real yields in the US for example). Real yields are indeed exceptionally low right now. Take index-linked gilts for example – real yields on the 2.5% 2013s are -2.1%, which means that if you purchase this security and hold to maturity then you are guaranteed to underperform UK RPI inflation by an annualised 2.1%. In the UK, you have to go down the yield curve to 2020 before you have a positive real yield. The same can be seen in US TIPS, where the 0.625% 2013 has a real yield of -1.8%, although European real yields are a bit higher (the German 2013 inflation linked government bond has a real yield of -0.2%, a reflection of lower anticipated inflation rates and a reflection of the ECB’s decision to hike nominal interest rates in the face of a perceived inflation shock).
PIMCO’s Bill Gross also highlighted Reinhart and Sbrancia’s paper in his recent investment outlook, and used this argument to explain why investors in developed market government bonds are being short changed. But I think he misses out on a very important point*, which is all to do with the shape of the yield curve. Let me explain.
Yield curves in the UK and the US in particular are exceptionally steep. Looking at the UK, the MPC’s decision to keep the bank rate at 0.5% is keeping the front end of the gilt market very well anchored. Short dated gilts are also being supported by heavy buying from relatively price insensitive central banks (all these FX reserves from around the world have to go somewhere – the Economist had an interesting take on this topic recently here). The yield on the UKT 4.5% Mar 2013 is just 1.0%, which isn’t particularly enticing.
However, once you go out a few years, the gilt yield curve starts steepening sharply and gilts look more interesting. The UKT 2% 2016 has a yield of 2.2%, while the UKT 3.75% 2021 has a yield of 3.5%. Many argue that an average annualised 2.2% total return for a five year gilt held to maturity is hardly a screaming buy, and nor is a 3.5% annual return from a 10 year gilt. But you also need to consider the effect of what’s called the ‘roll down’. This means that what is currently a five year gilt yielding 2.2% will ‘roll down’ the yield curve and will become a three year gilt yielding just over 1% in a couple of years, if you make the assumption that the yield curve remains exactly the same shape. And as we all know, when bond yields fall, prices rise.
So if the yield curve stays the same shape, then the drop in the gilt’s yield as it rolls down the curve will result in a capital gain on top of the income received from the gilt. To demonstrate this, firstly consider the UKT 4.5% 2013, which matures in March 2013. The yield on this gilt is currently 1.0%, so an investor buying this today will have had a total return of 1.0%pa at maturity. Not great. But assuming the yield curve doesn’t change shape between now and March 2013, the UKT 2% 2016 will have a yield of 1.4% in a bit under two years, so as well as clipping the 2% coupon each year, you’d also get a capital gain of 1.3%pa, providing a total return of 3.3%pa. This is a very important difference as a 3.3% annual return (coupon plus capital gain) is considerably more attractive than 2.2% (the bond’s current yield). In fact, the UK bond market’s five year inflation expectations for the RPI measure is 2.8%pa, which equates to inflation expectations of roughly 2%pa for CPI. If the UK yield curve remains the same shape, then the UKT 2% 2016 would give a return 1.3% in excess of current inflation expectations.
Using the same yield curve assumption, by March 2013 the UKT 3.75% 2021 will have shortened from being a ten year gilt to an eight year gilt, and its yield will have fallen from 3.5% to just above 3.2%. The income from the gilt plus the capital gain as the yield falls would give you a total return of about 4.7%pa over the next couple of years, which is again significantly above the headline yield of 3.5%, providing a decent positive real return.
Therefore, if you believe that the yield curve will remain steep and short rates will remain anchored (perhaps due to the massive public and private sector debt burdens, the vulnerable housing market and UK banking system, the likelihood of lower inflation next year, and the fact the UK economy is no bigger than it was six months ago), then you’d likely see government bond returns in excess of inflation, at least in the 5-10 year maturities where the yield curve’s still steep. Of course, if your view is that the Bank of England or Federal Reserve will start hiking rates then you’ll have a different view about the future shape of the yield curve and therefore the attractiveness of government bonds. Indeed, the market is pricing in a normalising of interest rates (there are two 0.25% rate hikes priced into the UK bond market by this time next year) and the ‘no arbitrage’ rule dictates that the yield curve is expected to steadily flatten over the next few years. But we disagree with what the market’s pricing in – we think rates will stay lower for longer, and the yield curve will remain steep. Given this view, the roll down effect illustrates why we’re uncomfortable getting too short duration, and most of the funds we manage are either marginally short duration or neutral. It’s expensive to be very short duration right now.
Congratulations if you’re still with me; here is your reward. Carmen Reinhart signed a copy of This Time is Different for me when I met her last year, and it’s such a good book that we’ve previously offered it to readers on this blog. The person who is closest to guessing where Spain 5y CDS is at the close of Tuesday 31st May wins (we’re taking the GCDS page on Bloomberg). To give you a guide, it briefly exceeded 350bps at the beginning of this year before rallying to below 200bps at the beginning of April, but continuing Eurozone sovereign debt fears have meant it’s since widened to back above 250bps.
Click here to email your entry. Click here to read and print off competition terms and conditions. All entries to be received by midnight Friday 13th May. The information we collect from you is used solely to contact you in the event that you have won and the winners name may be publicised.
* Note that it’s not the first time that we’ve disagreed with Bill Gross – see Richard’s response to PIMCO’s ‘bed of nitroglycerine’ comments here