Danny Walker, Dong Lou, Gabor Pinter and Semih Üslü
Government bond yields tend to drift higher in the days before monetary policy or data news in the UK. Over the past two decades this tendency – which we label ‘pre-news drift’ – has pushed up on yields by 2 percentage points in total over that period. The drift concentrates in pre-news periods that coincide with the issuance of UK government bonds, which is more common than it used to be. Our analysis shows that dealers and hedge funds are reluctant to buy bonds when news is on its way, which pushes up yields. Pre-news drift could affect the signal monetary policy makers draw from market rates and it could have implications for the optimal timing of bond issuance. There are further details in an associated working paper.
Bond yields drift higher when monetary policy announcements or macroeconomic data news are anticipated
Bond yields move around when news is released, especially news that tells investors something about monetary policy or the economy. That won’t be surprising to anyone. It’s much less obvious that yields should move in any particular direction before the news has arrived. But that is exactly what we observe in the historical data for the UK: a tendency that we label ‘pre-news drift’.
We analyse UK government bond yields since Bank of England independence in 1997 and focus on periods in the lead up to monetary policy announcements by the Monetary Policy Committee and inflation and labour market data releases by the ONS – both of which are scheduled releases, and therefore anticipated by the market.
Chart 1 shows that bond yields tend to drift upwards on average in the two days before the news is released. They tend to drift down on days that are outside these windows. This pre-news drift isn’t small – it has pushed up on yields by 2 percentage points, which compares to a total fall of 6–7 percentage points since 1997. We show in the paper that the drift has been larger at longer maturities and is mostly accounted for by risk premia, rather than monetary policy expectations. In this blog we set out our analysis of what is behind the drift.
Chart 1: Bond yields tend to rise in the two days before monetary policy announcements or macroeconomic data releases in the UK, and they fall outside of those periods
Sources: Bank of England and ONS.
The pre-news drift is concentrated in periods after the DMO has issued government bonds
Pre-news drift appears to be related to the timing of the issuance of UK government bonds – known as gilts – by the Debt Management Office (DMO). The pre-news drift shown in Chart 1 almost entirely occurs in pre-news windows that are preceded by bond issuance. This could mean, for example, windows where bond issuance on a Tuesday is followed either by a labour market data release on Wednesday or a monetary policy announcement on Thursday (or both).
Simple regressions detailed in the paper – which control for other relevant factors – reveal that the daily change in 10 and 20-year UK yields during pre-news windows is around 0.5 basis points larger on average than yield changes outside pre-news windows. That can be thought of as a baseline. But this difference rises as high as 1.1 basis points when the pre-news window is preceded by bond issuance. Cumulating these small differences in daily yield changes over multiple years explains the much larger 2 percentage point upwards drift highlighted in Chart 1.
You might expect the pre-news drift to unwind after the news has arrived – otherwise yields would remain permanently higher. We use regressions to analyse this in the paper and find no clear evidence for a reversal of the pre-news drift over the following couple of days. But of course this doesn’t rule out it happening. A reversal would be consistent with the downwards drift in other periods shown in Chart 1.
Events where monetary policy and macroeconomic data news coincide with bond issuance are increasingly common
The issuance of UK government debt has increased in the past couple of decades – gross issuance averaged £26 billion in the 1990s and £140 billion in the 2010s. This has meant that there are more and more periods where monetary policy announcements and macroeconomic data releases are preceded by DMO issuance: the pre-news windows in our analysis.
Chart 2 shows that in some years every monetary policy announcement was preceded by DMO bond issuance. These trends mean that the pre-news drift that we describe in this blog has been – and will likely remain – an important feature of the bond market.
Chart 2: It has become increasingly common for monetary policy decisions or macroeconomic data releases to be preceded by UK government bond issuance less than two days earlier
Sources: Bank of England, DMO and ONS.
The limited intermediation capacity of dealers and hedge funds explains the drift
Why does pre-news drift happen? Our analysis shows that the answer comes down to the behaviour of financial intermediaries like investment banks – or dealers – and hedge funds. Recent work has highlighted the importance of balance sheet constraints for bond market dynamics. Dealers in the UK bond market are responsible for buying a share of newly issued government bonds and then they tend to sell them on to other investors. But we use transaction-level data to show that when news is anticipated, dealers don’t want to be stuck holding too many bonds in case the news causes sudden price drops. As a result, they rush to sell the bonds they buy in the auction, which pushes prices down and drives yields up.
Our analysis also points to a shift in who steps in to buy bonds during pre-news periods. Hedge funds, which are usually active buyers, tend to pull back because they want to avoid committing to positions that might need reversing closer to the news event. In their place, less active investors – such as pension funds – take on a bigger role. These investors help maintain market liquidity, but they also demand higher returns for taking on risk during these periods, again driving yields up.
Is pre-news drift just a UK thing?
There is a temptation to assume that the pre-news drift reveals something unique about the UK’s government bond market. Indeed, a recent study showed that yields drift in the opposite direction – downwards – around Federal Reserve (FOMC) monetary policy announcements in the US, which is inconsistent with our analysis at face value. However, Chart 3 shows that if we focus on FOMC announcements that are preceded by issuance of US treasuries – the blue line in the chart – yields drift upwards, much like they do in the UK. This suggests that similar market dynamics might be at play in the US as well.
Chart 3: While US bond yields tend to fall around Federal monetary policy announcements, they rise when those announcements are preceded by issuance of US treasuries
Source: Bank of England.
Policy implications
Pre-news drift has implications for monetary and fiscal policymakers and reveals a narrow channel through which monetary and fiscal policy interact. First, pre-news drift could pollute the signal that monetary policy makers should draw from market interest rates. The use of averaging periods for market rates helps avoid undue emphasis on short-term volatility in rates. But if it persists, the drift might create a need for policymakers to alter their policy decisions to offset changes in market rates that don’t reflect the economic outlook. Consistent with this, in our paper we show that pre-news drift tends to be followed by bigger moves in market rates around monetary policy announcements. Second, there are implications for how issuers determine the socially optimal timing of government bond issuance – it might make sense for them to avoid periods where there is impending news.
Danny Walker works in the Governors’ offices, Dong Lou is professor at London School of Economics and HKUST, Gabor Pinter is an economist at the Bank for International Settlements and Semih Üslü is a professor at Johns Hopkins University.
If you want to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.
Share the post “Why do government bond yields drift when news is on its way?”
