Archive for November, 2016
“Mr Speaker, I am abolishing the Autumn Statement.”
2016 has already delivered much in the way of political upheaval. But no one was expecting this! Granted, there was much laughter as the new Chancellor told MPs that, having done away with the traditional spring Budget and Autumn Statement, he would be moving to an autumn Budget and a Spring Statement. (He had to clarify that this really was a significant change because the Spring Statement wouldn’t actually say much.)
Thus concluded the only truly dependable projection from HM Treasury. What did the rest of this, the last ever Autumn Statement, have to tell us?
Reports of the event were dominated by the forecast effects of Brexit. Of course, nobody knows what these will actually be. The Office for Budget Responsibility plumped for lower growth, higher inflation and more borrowing than before. Its guesses are worth no more than the Bank of England’s – though they were, interestingly, very different. The Bank, in its Inflation Report out earlier this month, had GDP for 2018 and 2019 coming in at +1.5% and +1.6%. The OBR thinks the answer will be +1.7% and +2.1%. Counter-intuitively the OBR is also forecasting lower CPI inflation (2.5% and 2.1% as against 2.7% and 2.5% over on Threadneedle Street). It is likely mere coincidence that higher growth and lower inflation will have combined to produce a less alarming debt forecast. In any event, the inconsistency serves to demonstrate nicely the speculative nature of all these kinds of exercise.
Regular readers will know how important the level of gilt yields is too. They have fallen since the Budget and this has delivered a projected bonanza of £24bn over the course of the next five years. That is the impact of a fall in gilt yields of all of 0.3%. (So long as they don’t go up again, the nation’s finances are safe. Well, safeish.)
The Chancellor briefly alluded to longevity and suggested that the next Parliament consider doing something about it. This is a good idea: the state pension bill is forecast to reach £101.9bn in fiscal 2020-21, up from £91.5bn this year. And that represents a compound annual increase of only 2.7%. More aggressive inflation, or a higher rate of earnings growth would push the cost higher.
We have no idea what growth and inflation will be over the next few years. Nor does the Treasury have much, if any, control over interest rates. There are, however, material numbers over which it does have some say. And in this febrile political environment, with Brexit on the horizon (or possibly not), public debt already forecast to reach 90% of GDP soon and no medium-term prospect of balancing the books, what is the expected net impact of the Chancellor’s policy decisions? Why, to widen the budget deficit by £4bn next year, £6bn in 2018-19 and £8bn the year after that! And those numbers are predicated on the basis of 43 discrete policy changes covering everything from tax reliefs available to museums to the exemptions regime for social sector rent downrating. Taxes up, borrowing up and the tax code as complex as ever. The ghost of Gordon Brown still hovers over the building whose magnificent remodelling he instigated all those years ago.
But – as many have felt, at times, this year – enough of politics. Despite the mainly Brexit-related fuss it caused the Autumn Statement contained nothing to suggest that the management of the country’s economy is to undergo change. The UK’s national debt remains problematic, and the Treasury seemingly content to be at the mercy of future events. That, from a pure investment perspective, is the key message from the last Autumn Statement in history.
The election of Donald Trump to the US presidency apparently “wiped out more than $1 trillion across global bond markets“, as reported by Reuters earlier in the week. The inflationary nature of his policies has not, after all, gone unnoticed. Bond markets have digested new information and responded rationally, by falling in price so as to offer investors the prospect of higher real returns. Yes, the politics have tinged the reporting – as with the economic consequences of the UK’s decision to leave the EU. But surely the picture is clear: an event has occurred, markets have responded efficiently and their response is cause for concern.
As in the case of Brexit, however, the political element seems to have cost some observers their perspective. Starting with the obvious: only part of that $1trn comes from the US itself. And bond markets have been weakening for some time: Mr Trump’s election only accelerated the process. Referring to the BofA Merrill Lynch US Treasury Index, the full market value of US government bonds climbed to a peak of $9,729bn on the 8th of July, when the ten year bond yield reached a record low of 1.36%. By 8 November, before there was any indication of the surprise election result, this value had already fallen to $9,507bn. As of yesterday, the number was $9,296bn. So most of the fall since the summer preceded the new president entirely.
More broadly the invocation of a “Trump thump” is a symptom of the behavioural concept known as anchoring. We had become very used to both the recent level of bond yields and the pace of their rise before the election, so the sudden change shocked us into thinking that something anomalous had occurred: in this case, a dramatic change in policy direction arising from the victory of a candidate from left field.
This is at best a partially misleading analysis. Inflationary pressure has been mounting – and disinflationary forces have been receding – for some time. Look at the prices of base metals and freight and global activity seems to have been picking up over the second half of this year too. What is strange – perhaps – is that bond yields remained at record lows for such a long time. Putting it unkindly, bond markets have often seemed to care much more about the last ten weeks than the next ten years. There is a case to be made that the so-called “thump” was no more than a catalyst to their ongoing recovery from inertia.
Time for some more context. Yes, the US 30-year yield has risen by its fastest pace since at least 2009, putting on 40bp in four trading sessions starting last Wednesday week. But it has since spent the whole of this week hovering around the 3% mark. The initial spike was unusually volatile, but taking a slightly longer view the 30-year yield has now risen by about 1% in total since its July low – and did exactly the same thing over a similar period in H1 2015.
Look at current pricing in absolute terms and 3% is not even a high number. Trump has only managed to thump the 30-year Treasury yield back up to where it ended last year. It is almost exactly in line with its five-year average. And this is still miles below the 5.3% it posted on the cusp of the credit crunch (21 July 2007).
From another angle, a 3% return for thirty years looks plausible in real terms if we focus on the most recent American CPI data (last in at +1.6%). But long run expectations according to the University of Michigan survey are closer to +3%, and the 21st-century average prior to the Great Recession was +2.8%. On that basis the election result has only delivered US investors a prospective 30-year real return of 0.0-0.2%. In fact if one really wanted to get bearish on bonds one might observe that these levels of inflation would, not so very long ago, have been regarded as unachievably benign in any kind of growth-positive environment for the US economy. That is why the bond market has been able to rally to new highs year, after year, after year since inflation peaked at 15% in 1980.
So from a politically detached perspective, perhaps we should forget about Mr Trump’s impact on the US bond market. Fundamentally speaking he has taken the stage as accomplice, not assassin.
What we oughtn’t to do, of course, is forget about the possible impact of bond markets . . .
Well that clinches it, surely. Today’s US data showed the unemployment rate down again to 4.9%. Payroll growth continued at a strong pace, with non-farm jobs up by 161,000 last month. Both these prints came out in line with expectations – but wage growth beat every estimate going, hitting a new post-recession high of +2.8% on the year. Purchasing manager surveys out over the last couple of months suggest that the rate of job creation will accelerate into the end of this year if anything. The American labour market is showing clear signs of warmth. Surely, a Fed hike next month is certain.
This chimes with the consensus view. Of the 66 forecasts currently made available to Bloomberg, 15 expect no change and all the rest are for a 25bp hike. The interest rate markets also expect the Fed to see out 2016 with a target rate of 0.75%, then go on to hike again around the middle of next year.
So far, so uncontentious. Indeed it is reassuring from an investor’s perspective that the market reaction to somewhat firmer expectations for interest rates has, thus far, been sanguine. Halfway through this year a December hike had yet to be priced in; by the autumn the eurodollar futures market had become unequivocal on the subject – Treasuries sold off too – but the S&P 500 still rallied, turning in its strongest quarterly performance of the year so far.
The key word there is: “somewhat”. In the US as in Britain there is now a real risk that tightening occurs more quickly than people think.
That +2.8% wage growth is part of the reason why. A connected reason is that the energy sector has recovered some of its strength this year, taking away a recent source of downward pressure on activity and employment. Then there are import prices: stable oil and a stable dollar have now closed off that source of deflation. Core CPI has already been running at its strongest sustained level this year since the Great Recession. Both the headline CPI measure and the consumption deflator used to calculate chain-weighted GDP have been catching up. Finally, unreliable indicator as it is and completely unfashionable as it has become, broad (M2) money supply growth is running at its fastest pace for nearly four years.
The question to which nobody knows the answer is: will the labour market start really overheating and, in the circumstances, contribute to an uncomfortable level of inflation? At the moment, US policymakers are split on the subject, but the Fed’s actual monetary response to date, together with the interest rate markets’ pricing, implies a near-total lack of concern.
Another wild card is our old friend political risk. At present, polling at the national level and in some of America’s “battleground states” suggests that Secretary Clinton is on course for a narrow victory. If Mr Trump were to pull off a surprise win, however, it is not just the market response but the economic consequences which could be significant. There has already been extensive coverage of the effect that the candidates’ fiscal policies might have on the national debt, but consider some of the Republican candidate’s other measures: the expulsion of migrant labour, swingeing tariffs on imports, less accommodative trade agreements (as well as big tax cuts for businesses and households). All of these would be inflationary.
It would be imprudent to expect a crisis. But the possibility of a problem is clear and what matters to us as investors is that this is not recognised – indeed, quite the reverse. Bond markets think that inflation over the next ten years will run at an average of 1.7%, below the 2.1% they have priced in on that horizon over the last 15 years. Consumer expectations for near term inflation, as measured by the University of Michigan survey, are below average too, and for long term price behaviour are at their lowest ever level.
So there could well be a surprise or two in store. And whatever its scale, this ought to have consequences for portfolio construction.