It crops up as a topic of discussion every now and again: the ‘term premium’. But what is it exactly? Simply put, the term premium is the bonus that investors traditionally received for the risk of owning longer-dated bonds – the duration risk. And yes, that’s ‘received’, past tense. In the last decade of global central bank quantitative easing, bond term premiums have been compressed as never before, pushing sovereign yields down worldwide.
A 10-year bond yield can basically be broken down into two main components: the average future short-term interest rate expected over a 10-year horizon and a term premium. The first is largely dependent on local monetary policy expectations, while the second appears to be more affected by global developments. As unconventional monetary policies are now set to normalize – starting in the US – what will be the impact on the term premiums and will this produce a synchronized global upswing in bond yields?
While it may seem logical to assume that long-term yields are mainly driven by local monetary policy, the term premium component can in fact sometimes exert a much stronger effect on yield dynamics. The synchronicity of US, German and Japanese term premiums in recent years has been striking and suggests that common global forces are present and affecting these markets, even though monetary policy on a local level may be quite different. One obvious culprit is the global shortage of safe assets – initiated by the rapid and substantial increase in central bank foreign exchange reserves since 2000 and exacerbated by quantitative easing.
This trend is confirmed by the fact that, prior to QE, common global factors accounted for about 69% of the changes in individual term premiums, a percentage which jumped to 83% at the end of February this year. Because central banks such as the ECB and the Bank of Japan have bought substantial amounts of sovereign bonds in their domestic markets, international bond investors have been forced to shift their portfolios to regions where bonds are less scarce or into riskier assets. This portfolio rebalancing has been responsible for the co-movements in term premiums in developed markets. For example, the outflows from Japan in 2014 and the Eurozone in 2016-2017 have been – at least partly – invested in US Treasuries, contributing to downward pressure on the US term premium and depressing US long-term yields.
So, although the decline in the US term premium in 2017 and its move into negative territory may seem counter-intuitive given that the Fed had already started to shrink its balance sheet, it illustrates the extent to which global factors affect individual term premiums. There is nothing to suggest that this situation will change in the immediate future. Local differences in terms of the scarcity of domestic bonds may well affect the pace of normalization, but are unlikely to become a decisive factor.
Japan, for example, is at one extreme end of the spectrum. Its government has such a tight grip on the market that trading volumes have evaporated. A bizarre record was set on 13 March 2018, when not a single 10-year benchmark note was traded on the Tokyo Stock Exchange. The German Bund free float is not that abundant either – no more than 10% according to an ECB analysis, and in stark contrast to the 50% estimated for US Treasuries. All else being equal, a higher free float may lead to a more pronounced change in term premiums, so the US should be the first country to see some meaningful adjustment.
The gap between current term premiums and their pre-QE averages is substantial. That said, a sudden major correction, driving bond yields higher, is unlikely in the coming quarters. We think there will only be a mild upswing of no more than 25 basis points for US Treasuries and 10 basis points for Bunds. This may sound rather modest, but global central bank bond holdings will only start to decline toward the end of 2018 and this is likely to be a very gradual process.
Interest rates in both the US and Germany will probably rise to a certain extent. But don’t expect rising term premiums to be a major contributory factor in this process. In the current phase of the cycle and with official bond holdings still at high levels, any such rise in yields is more likely to stem from a re-pricing of central bank rates than from a jump in the term premium. For the foreseeable future, the risk associated with holding long-dated paper will continue to be rather poorly rewarded.