Fed – Sense of déjà vu

Share Button

Recent US data have likely put a Fed rate hike at its 21st September policy meeting beyond reach, with a post-US presidential election rate hike now the more feasible scenario.

US labour market data for August – sandwiched between very weak ISM prints – suggest that there is still slack in the US labour market. Aggregate weekly payrolls in the private sector rose only 3.5% year-on-year, which in turn is likely taming inflation.

Job creation growth was stable at around 1.9% yoy, which in itself is compatible with a Fed hike in September. But working hours fell in August, earnings growth slowed further and the pool of potentially available workers has now increased for three consecutive months.

These numbers will have done little to convince Chairperson Yellen that the time for reflection is over and the market is now only pricing rate hikes of 5bp and 13bp, respectively, for the Fed’s September and December meetings.

There is a sense of déjà-vu and the path of least resistance is probably for the Fed to keep rates unchanged this month while keeping alive the possibility of a hike on 14th December.

Moreover, the Fed will have the added benefit in December of knowing the still uncertain outcome of the US presidential elections scheduled for 8th November.

The Fed remains apolitical and has in the past hiked rates in the run-up to US presidential elections, but a Clinton victory and a positive US and global market reaction are probably necessary if insufficient conditions for the Fed to hike with conviction.

Should the Fed hike in December, this almost unprecedented glacial pace of rate hikes would be in line with my January forecast of only one or two hikes in 2016.

Financial markets’ reaction to recent US data has been a familiar one – lower rates, weaker dollar and stronger EM currencies and global equities. But it has revealed little about how markets are likely to respond if and when the next Fed rate hike comes into clearer focus.

September rate hike unlikely after lukewarm US labour market data

In Is it a hawk, is it a dove…not it’s the Fed (1 September 2016) I argued that the case for a US rate hike, based on current US and global indicators, was in aggregate as strong as it was when the US Federal Reserve (the Fed) hiked its policy rate 25bp in December 2015. But I also made the point that a strong set of August labour data was a necessary if not sufficient condition for a fundamentally cautious Fed to hike 25bp at its 21st September policy meeting.

As is often the case, Friday’s labour market data – examined in more detail below – had something for everyone across the hawkish-dovish scale. But the conclusion which I draw is that there is still slack in the US labour market, which is taming growth in aggregate weekly payrolls and in turn inflation.

The latest set of labour data – sandwiched between very weak manufacturing and non-manufacturing ISM prints – revealed that job creation growth was stable in August but that working hours fell and earnings growth slowed. These numbers will have done little to convince the more dovish Fed members, including Chairperson Janet Yellen, that the time for reflection is over and the market is now only pricing rate hikes of 5bp and 13bp, respectively, for the September and December meetings.

There is a clear sense of déjà-vu, with US data have once again likely fallen short of the rate-hike-hurdle which the Fed has seemingly set itself. So barring surprisingly strong economic data in the next fortnight and a sharp repricing of the rates market’s expectations, the path of least resistance is probably for the Fed to stay on hold this month while keeping alive the possibility of a rate hike on 14th December.


Path of least regret – Wait for still unpredictable outcome of US presidential elections

Moreover, the Fed will have the added benefit in December of knowing the outcome of the presidential elections scheduled for 8th November. At present, it is probably too close to call, with Hilary Clinton leading by only 2-5 percentage points in most opinion polls. And if there is one lesson to take from the recent UK referendum on EU membership it is that the will of the people is fickle and hard to predict.

The Fed is an apolitical body and would thus never acknowledge, at least not explicitly, that any future interest rate changes may be conditional on a specific political outcome. It has after all shown a willingness to look past discrete political events, hiking its policy rate in the run-up to both the 7th November 2000 and in particular the 2nd November 2004 presidential elections (see Figure 4).

On these two occasions, however, the presidential candidates – incumbent Texas Governor George W Bush, incumbent Vice President Al Gore and Senator John Kerry – were well-known quantities with reasonably predictable economic programs. In stark contrast, Republican presidential candidate Donald Trump has little political history. He has suggested economic policies which go well beyond the “normal” Democrat-Republican policy spectrum, which if enacted may have serious implications for both the US and global economy, not to mention financial markets.


Clinton victory – Necessary if not sufficient condition for December Fed hike?

In comparison, Democratic presidential candidate Hillary Clinton is a known quantity, having been Secretary of State for four years. She is more likely to pursue orthodox economic policies palatable to financial markets than Trump. Therefore, should she become the 45th US President, US and global equity markets are more likely to react positively, in my view, or at least take her victory in their stride.

This sequence of events would probably remove one major obstacle to a December rate hike, even if the US dollar – range-bound since early May – also appreciates somewhat (see Figure 12). With this in mind, the Fed’s path of least regret is probably to wait until December and hike rates with a greater degree of conviction, rather than hike in September only to backtrack because Trump has been elected President and sent US and global markets in a tailspin.

Should the Fed hike in December, this glacial pace of rate hikes would be almost unprecedented in recent times but in line with my January forecast of only one or two hikes in 2016 (see what if the Fed hikes, leaves rates on hold…or cuts, 29 January 2016). In the past 25 years, only once has the Fed delivered a one-off hike before staying on hold for a significant period of time. On 25 March 1997 the Fed hiked rates 25bp to 5.50% but then left rates unchanged for 17 months, before delivering three successive rate cuts in August-November 1998 at the height of the Russian crisis (see Figure 4).

Olivier Desbarres Fed Deja Vu Fig1

Growth in aggregate weekly payrolls at its lowest since end-2013

The Bureau of Labour Statistics publishes a vast and detailed set of data but the nominal index of aggregate weekly payrolls for nonfarm private sector employees perhaps best encapsulates the relative strength of the US labour market, in my view. As the product of average hourly earnings (in current $-terms), average weekly hours and employment this index captures many quantitative and qualitative aspects of the US labour market which neither payrolls nor the unemployment rate do.

Specifically, the index is a good gage of underlying economic activity and importantly whether the labour market is likely to drive consumption and in turn inflation. In comparison, headline non-farm payrolls and the unemployment rate are, taken in isolation, poor measures of the relative tightness of the labour market and its inflationary potential.

Aggregate weekly payrolls flat-lined in August, resulting in the year-on-year (yoy) growth rate slowing sharply in to 3.5% – the lowest rate since December 2013 (see figure 1). With data only going back to March 2007 it is difficult to conclude whether this growth rate is compatible with Fed rate hikes. However, the very rapid and sustained slowdown in aggregate payrolls growth from around 5% yoy in late-2007 to
-5% yoy in mid-2009 correlated with the Fed’s 512.5bp rate cutting cycle. In comparison, when the Fed hiked rates in December 2015, payrolls growth was running at a reasonably steady 4.5% yoy – a full percentage point higher than the current growth rate.

Quite clearly the Fed takes into account a far broader range of variables when setting policy rates, but it is conceivable that the Fed would at the very least want to see the growth rate in aggregate weekly payrolls stabilising before it countenances a rate hike. Let us look in turn at the three underlying components of aggregate weekly payrolls.

Olivier Desbarres Fed Deja Vu Fig2and3

  • The private sector added 589,000 new jobs in June-August, with job creation growth unchanged in August at 1.9% yoy in August (see Figure 2). Based on precedent, this in itself is not incompatible with a modest rate hiking cycle (see Figure 4), even if job-creation hit 742,000 in the three months prior the December 2015 rate hike. In contrast, surprisingly weak non-farm payrolls in May of +24,000 (- 1,000 in the private sector) were likely a key contributor to the Fed keeping rates on hold at its June 2016 policy meeting.

Olivier Desbarres Fed Deja Vu Fig4

  • But growth in hourly earnings in the non-farm private sector dropped to 2.4% yoy in August from 2.7% yoy in July (see Figure 3). The slowdown in earnings growth was seemingly due to both the relative increase in workers employed in less well paid sectors (e.g. food services) and workers generally having to accept more modest increases in hourly earnings.

Olivier Desbarres Fed Deja Vu Fig5and6

  • More significantly, the number of weekly hours worked fell in August to for the second time this year (see Figure 5), resulting in a further acceleration in the year-on-year rate of contraction to a six-and-a-half year of low -0.9% (see Figure 6). This fall in working hours occurred despite a sizeable increase in full-time jobs (and more modest decrease in part-time jobs) and increase in the share of full-time workers to an 8-year high of 82% (see Figure 7). This implies that the number of hours worked in part-time and/or full-time jobs decreased in August.

Olivier Desbarres Fed Deja Vu Fig7and8

Slack in labour market doesn’t point to rampant wage-price inflation

In a nutshell, the number of people joining the private sector continued to grow at a decent pace in August but existing and new workers were employed for fewer hours and commanding more modest gains in earnings. This suggests that the overall demand for labour remains modest, particularly in better-paid sectors, while the supply of potential workers remains decent – a point I made back in May in US Economy Not At Full Employment (13 May 2016).

Indeed, while the share of the labour force actively seeking a job – the much focussed on unemployment rate – remained near multi-year lows in August (see Figure 8), the pool of potentially available workers has increased in recent months. Figure 9 shows that the total of those unemployed and not in the labour force but currently wanting a job rose 323,000 in June-August, albeit from a low base. This was the first time since late-2012 that three consecutive monthly increases have been registered. As a result their share of the working age population edged higher to 5.4% in August from 5.3% in May.

Olivier Desbarres Fed Deja Vu Fig9and10


To recap, the (nominal) purchasing power of the private sector derived from regular weekly earnings (i.e. excluding other sources of income) stagnated in August and in year-on-year terms was up only 3.5%. This growth rate may be insufficient to drive measures of US inflation which have stagnated in recent months (see Figure 10). In real-terms – using core PCE-inflation as the deflator – aggregate weekly payrolls growth was only 1.9% yoy in August – the slowest rate in nearly six years (see Figure 11).

If the US labour market was very tight, let alone the economy nearing full-employment or over-heating, the growth rate of the working population would likely, somewhat counter-intuitively, start to moderate from high levels as employers would struggle to find new workers. At the same time, hours worked and earnings would likely rise rapidly as employers would have to pay more to retain existing workers and/or attract new workers. Strong growth in aggregate payrolls would help drive consumption and in turn inflation, putting further pressure on the demand for labour and wages and creating further inflationary pressures.

This was arguably the case in mid-2007. Job creation had slowed to around 1.3% yoy, but earnings growth was running at a robust around 3.6% yoy and the number of hours worked was ticking up 0.3% yoy. This wage-inflation spiral is arguably not in play in the US at present.

Olivier Desbarres Fed Deja Vu Fig11and12

Financial markets comforted by goldilocks data…for now

The reaction of financial markets to recent US data has been a familiar one in the past four trading sessions. US financial markets, perhaps unsurprisingly, have reacted dovishly to the tepid US labour market data and outright weak ISM numbers. The rates market has cut back its pricing of Fed hikes by about 2-3bps for both the September and December meetings and the US dollar nominal effective exchange rate is down about 1.2% since the middle of last week (see Figure 12).

The market’s belief that a tightening of US interest rate policy is unlikely before December while the US economy remains reasonably strong has once again boosted global risk appetite which was treading water in August. Global equity markets have inched 1.4% higher and a GDP-weighted basket of emerging market currencies (excluding the Chinese renminbi) has appreciated 1.6% versus the US dollar (see Figure 12).

This price action reveals little about how markets are likely to respond if and when the next Fed rate hike comes into clearer focus. The Fed’s December rate hike ushered in a multi-month fall in global risk, with markets concerned about the prospect of a more rapid pace of US policy rate hikes. The Fed will clearly want to avoid a repeat of such a scenario and will most likely continue to emphasise its well-known stance that future rate hikes are likely to be modest and gradual.

The credibility of this message is likely to be conditional on the US economy remaining just short of strong. Put differently, in order for global risk appetite to improve further, US data in coming months will need to be sufficiently weak to keep rapid rate hikes at bay, but sufficiently strong to reassure markets that the world’s largest economy and consumer is underpinning global growth and demand – a topic I will discuss in greater detail in my next research note.

Olivier Desbarres

Olivier Desbarres currently works as an independent commentator on G10 and Emerging Markets. He has over 15 years’ experience with two of the world’s largest investment banks as an emerging markets economist, rates and currency strategist.

Share Button