If you read a lot of Street research (especially of the macro variety) you know that in many cases you get to the end and are left wondering if the analyst actually said anything worth saying.
That’s not an attempt to belittle the analysis. It’s often quite good. It’s just that much like a lot of the social science research you’ll read in peer reviewed journals, it’s not always clear that all the math and mental acrobatics actually ends up telling us anything that wasn’t intuitive.
As far as macro research goes, the tie-breaker for me in those scenarios is usually whether the topic under discussion is something that’s being heavily debated by market participants. That is, if the topic is “top of mind” – so to speak – then any incremental information is probably worth highlighting.
So that’s the context for a Goldman note out Wednesday afternoon entitled “A Rough Awakening For Bond Bears” (why they went with “rough” there instead of “rude” is a mystery).
The note is about the rally in Treasurys and how we can try to understand it in the context of Japanese and German yields. Obviously this is something everyone is talking about now that 10Y yields in the US have come to proxy for “faith in reflation” (I mean, they always proxied for that, but now it’s not just rates and FX strategists talking about it).
So what I want to do here is present some excerpts below for you to consider with everything said above as the setup and allow you to determine for yourself whether Goldman said anything worth saying.
Not long ago we were often asked why our 2.5% end-March 2017 forecast for 10-year US Treasury yields was so low. Now, the most recurrent questions are how much further can bond yields fall, and what will it take to bring them back to the top of the Q1 range by end-June (as we continue to expect)?. This brief update attempts to answer these questions, drawing on our analytical toolkit and some judgmental observations.
Bond market returns tend to be highly correlated across the main advanced economies. This is largely because they are seen as relatively close substitutes by global investors, and long-term yields are moved around by a common ‘bond premium’. Using macro-econometric techniques, we can monitor which bond markets have been most influential in determining the recent price action, and how intermediate maturity yields currently relate to their macroeconomic underpinnings.
Exhibit 1 below plots the cumulative ‘shocks’ originating independently in each of the four major bond markets once cross-country ‘spill-over’ effects are taken into account. Since the ‘shocks’ are standardised, they can be interpreted as ‘forces’ pulling global yields down (when the line is negative) or pushing them up (when the line is positive).
In the big picture, US Treasuries have represented a bearish force for global fixed income since last June, and particularly since the election of President Trump. Japanese government bonds, by contrast, and more recently UK Gilts, have been pulling the rates complex in a dovish direction, albeit to a smaller extent. German yields have been the ‘tie-breaker’ in these dynamics.
Since the local highs in yields recorded in mid-March, our analytics suggest that German Bunds (and other European highly-rated bonds) have taken the upper hand, while US Treasuries have no longer had enough strength to offset this. This is as much as statistical analysis allows us to say: the Euro area is pulling global rates down, while the US is no longer pushing them up.
Moving on to macro-economics factors, Exhibit 2 plots the distance, measured in standard deviation units, of 10-year generic bond yields from an estimate of macro ‘fair value’ obtained through our Bond Sudoku framework. Our estimates account for the evolution over time of analysts’ expectations around future real GDP growth, CPI inflation and policy rates in both the market in question and the rest of the advanced economies. These expectations are collated by Consensus Economics and drawn on a monthly basis. The latest observation is April.
As the chart illustrates, until recently US Treasuries were an outlier: their macro valuation was closest to ‘fair’, while those of their European and Japanese peers were stretched (on the border between one and two sigma ‘expensive’). Across Europe and in Japan, bond yields remain largely influenced by the large-scale bond purchases (QE) conducted by central banks in an effort to push investors towards riskier assets and reflate the economy. These purchases have driven the bond premium deep into negative territory – influencing long bonds globally. At its widest point in March, the macro valuation gap between the US and the rest of the G-4 was in the 80th percentile since the 2008 financial crisis – a vivid illustration that the US was doing all the legwork for bond bears.
Combining the information gleaned from the statistical analysis with that from our modeling work associating nominal yields to expectations on macro factors, the impression is that US bonds had moved too far ahead of the rest of the pack in Q1, and needed continued affirmation from the domestic macro front of their new status, or have other markets catch up, to remain at that level or move even higher. In the event, investors have been disappointed on both fronts, and have had to pare back their underweight duration exposures as:
- In the US, more clouds have gathered over the size and the timing of the Trump Administration’s fiscal expansion, and the news on the inflation front has been more mixed of late
- In Japan, the BoJ is persevering with its current yield curve control policy, harnessing the entire nominal yield curve. Our Japan Economics team expects the central bank to reaffirm its short-term (negative 10bp) and 10-year (0%) interest rate targets at its upcoming policy meeting.
- In the Euro area, two developments have come to the forefront: First, weaker-than-expected headline and core inflation for March has further amplified the dovish communication from the ECB, thereby pushing back on policy normalisation expectations. Second, political risk has increased ahead of the first round of the Presidential elections in France, to be held on Sunday, 23 April.