The uncertain science of prediction
If you think the bond market has got it wrong – check your assumptions. So goes the conventional line of argument, and it is one that has been repeatedly thrown by economic commentators in the direction of monetary policymakers of late.
The leap in bond yields after Ben Bernanke floated the idea in May that the Federal Open Market Committee would consider tapering its asset purchase programme (if the data warranted it) has given rise to fevered discussion about an exit from current monetary policy. Much of it has focused on whether the bond market overreacted to Bernanke’s statement or if he was genuinely signalling unexpected policy tightening.
Such was the scale of the market shift that the FT’s traditionally sanguine Gavyn Davies asked whether the Fed had in fact lost control of short term interest rates. And he had good grounds for doing so - the expected short term rate by mid-2016 rose by around 100 basis points with markets pricing in a 30% chance of tightening next year. He mused [emphasis mine]:
“There is evidence that this signalling effect of Fed balance sheet changes might be very powerful. If the Fed is not willing to “put its money where its mouth is” by buying bonds, then the market might take its promises to hold short rates at zero less seriously than before. According to this recent research by the San Francisco Fed, it is possible that a sizeable proportion of the total effect of QE on bond yields came from these signalling effects rather than the portfolio balance effects which have usually been emphasised by the central banks.”
That is to say, talk alone may be sufficient to shift expectations about the trajectory of policy. This was certainly the line taken by The Economist’s Ryan Avent who argued in June that the Fed clearly did signal a departure from its previous stance – irrespective of whether they intended to or not:
“But it's frustrating that the message reversal, when it comes, will probably reflect thinking within the Fed that, "markets misinterpreted what we were saying, and so we had to set them straight". That's not what will have happened. Whether the Fed gets it or not, it set a tighter monetary policy last week. And if a new policy message calms markets it will be because the Fed is effectively setting a more accommodative policy than it did on the 19th.”
The reaction to Bernanke’s speech clearly did suggest that the market was looking through the QE taper talk at the implications it could have on the likelihood of rate rises. And here is where the Fed’s surprise is perhaps understandable. The statement was an attempt to guide expectations over a gradual wind down of its asset purchases, not the trajectory of interest rates.
To me, this seems like a crucial message to be taken from the taper discussion. Central banks have been looking closely at how to exit extraordinary measures, but few appear to have connected unwinding QE directly with interest rate expectations.
The first point to make is that the uptick in yields both strengthens the case for forward guidance and suggests that its effects at the zero lower bound are likely to be asymmetric. If the market is pricing in a policy tightening beyond what the central bank deems desirable (e.g. a yield spike that threatens to disrupt the pace of recovery) it becomes incumbent on monetary policymakers to make their case in public.
That said, because policymakers are understandably unwilling to commit to defined timetables for rate rises or tapering – including taking a flexible approach to the Evans rule unemployment trigger – the likelihood that the market will overreact to hints of tightening are high. If market participants cannot, by definition, know when a rate rise will occur they are bound to try and read the runes of speeches and committee minutes to identify any change in tone.
Given that under current policy central banks in the West cannot lower rates at the ZLB even ultra-dovish statements are unlikely to move expectations much unless they are matched with increased asset purchases. This, however, has the unfortunate consequence of tying the two policy levers together, so that when tapering is put on the table market participants assume rate rises are likely to follow and act accordingly.
(Note – Adam Posen has long warned of the limitations of forward guidance if central banks are seen as unwilling to “put their money where their mouth is”)
The takeaway here is that careless talk can cost a great deal to unwind.
So what does this mean about exiting current policy?
In a speech in September 2011 David Miles, a member of the Bank of England’s Monetary Policy Committee, argued in favour of raising the base rate before attempting to exit from asset purchase programmes:
“I believe there are likely to be advantages to raising Bank Rate first, and I would expect this to be the strategy. First, the costs of reversing any premature tightening would appear to be lower. And second, it may have advantages for financial stability because some banks and building societies have indexed a substantial proportion of their lending to Bank Rate, while their funding costs depend on market conditions. So keeping Bank Rate low squeezes the margins of these financial institutions and weakens them."
I think Miles is not only wrong here, but potentially dangerously so. He appears both to underestimate the potential impact of a rate shock on heavily indebted households and businesses and also the importance of low rates in protecting banks rather than weakening them.
Ultra low rates have been a huge boon to the banking industry as financial institutions attempt to rebuild their balance sheets. Low rates either automatically reduce debt servicing costs for floating rate borrowers or allows them to refinance at more favourable terms. This protects debtors, yes, but it also allows people will falling real incomes to continue to service their debt rather than defaulting – and therefore averts the need for large-scale writedowns of the value of the loans.
The idea that we should worry more about the profit margins of seriously damaged financial institutions, especially those in the UK, rather than the current quality of their balance sheets seems a little too optimistic to me.
Moreover, leading the policy exit with a rate rise may not be as easy as Miles imagines. You only need to look at the residual impact of taper-gate across government debt markets to see the difficulty in unwinding perceptions of a policy shift.
And here’s the problem if a rate rise goes wrong. The government forecasts that the UK housing market is set to undergo significant pain as economic conditions start to normalise. Under its base case scenario, using Office for Budget Responsibility forecasts, they predict:
“Annual possessions are 10.6 percent higher in 2013 than in 2011, and possessions numbers increase further in 2014 and 2015. Possessions are 38 percent higher in 2015 than in 2011, as interest rates, assumed to start rising from 2013Q3, are not sufficiently offset by falling unemployment and a slightly faster rise in disposable income and in house prices.”
On its own this would be an unpleasant experience for those pushed into negative equity over the crisis whose woes have been compounded by persistent falls in real incomes. Moreover, that report was written before the OBR’s downgrades to its outlook in December so the situation may have deteriorated since then.
Yet if the Bank of England were to lift rates from the floor prematurely the impact on the mortgage market could be even more severe. Below is a chart from the Bank’s Financial Stability Report from June showing the impact of interest rate rises on households:
As the report states a rise in interest rates, without a strengthening in income, could significantly increase borrower distress and losses to banks. One indication is that households accounting for 9% of mortgage debt would need to take some kind of action — such as cut essential spending, earn more income (for example, by working longer hours), or change mortgage — in order to afford their debt payments if interest rates were to rise by just 1 percentage point. This would rise to 20% of mortgage debt if interest rates were to rise by 2 percentage points.
The report also notes that, unlike the US where long-term fixed rate mortgages are the norm, the share of mortgages with fixed rates in the overall stock is close to historical lows in UK (although interestingly, there is some evidence that so-called adjustable-rate mortgages (ARMs) are becoming more prevalent in the US as well, potentially as a result of the low interest rate environment).
Due to this the damage of a premature rate rise (or the perception of a premature rate rise) to an already impaired financial system could be substantial, leaving aside the huge implications for families who find themselves without a home after having struggled to meet repayments through the crisis. What this stresses is the critical importance of setting out appropriate metrics for the Monetary Policy Committee to use in order to justify raising rates.
For me, the most sensible suggestion is for the central bank to target a price index that gives substantial weight to nominal wage growth. The case for this was set out in a 2002 working paper from the European Central Bank in which the authors conclude:
“Our results suggest that a central bank should give substantial weight to the growth in nominal wages when monitoring inflation. This conclusion follows from the fact that wages are more cyclically sensitive than most other prices in the economy (which is another way of stating the well-known fact that the real wage is procyclical). Moreover, compared to other cyclically sensitive prices, wages are not subject to large idiosyncratic shocks. Thus, if nominal wages are falling relative to other prices, it indicates a cyclical downturn, which in turn calls for more aggressive monetary expansion. Conversely, when wages are rising faster than other prices, targeting the stability price index requires tighter monetary policy than does conventional inflation targeting.”
Central banks may have good reasons to taper their asset purchase programmes in the near future (due to untoward effects of the policy or at the very least diminishing returns). The communications strategy for forward guidance should therefore stress that the metrics for gradually withdrawing QE may not equate to those that would warrant a rate rise. Failure to do so could have disastrous economic consequences.Without a sustained rise in incomes it is difficult to see how rates could responsibly be lifted from their current levels.