Published on January 29, 2016
1. 1 What if the Fed hikes, leaves rates on hold…or cuts? The future path of US interest rate policy remains a major source of uncertainty for global markets, with the rates market far more dovish than analysts and in particular FOMC members. Someone will ultimately be proven wrong and this will impact asset prices, including the US dollar. Previous hiking cycles suggest that macro and market conditions in the US and globally are not conducive to the US Federal Reserve hiking four times this year, as currently forecast by FOMC members. US and global GDP growth and headline inflation, as well as US manufacturing ISM, house price inflation and equity markets, are currentlyfar weaker than in the run-up to the 1994, 1999, 2004 and 2005 Fed hiking cycles. At the same time, resilient US and global core inflation and strong US non-farm payrolls currently argue against the Fed taking the unprecedented step of reversing its recent rate hike and in the process sinking its reputation. At this stage the Fed still looks more likely to muddle-through with a couple of hikes in 2016 – more aligned with current market pricing. The implications for the US dollar, which I examine in detail, are partly conditional on the respective performance of the US and other major economies and the perceived risk from tighter US policy to both domestic and global growth and confidence. In any case a central assumption is that a decoupling of the US economy from the global macro picture is veryunlikely. I still expect further modest US dollar appreciation to prevail – and accelerate if the worst-case scenario of a global recession and risk aversion materialises. But the dollar’s 18-month rally is at risk if tighter US monetary policy starts to weight more heavily on US macro indicators and the market perceives long gaps between Fed rate hikes as a tacit admission that the December hike was ill-conceived. The idea that the Fed finally hiking rates after nearly a decade would somehow quash market uncertainty was always wishful thinking (see The first hike is the hardest…and so is the second and third, 15 December 2015). The Fed’s interest rate path, along with China’s economic prospects, is still the elephant in the room and there is already a chasm between market, analyst and FOMC expectations (see Figure 1). In a survey I conducted in on 15 December (i.e. before the 16th December Fed hike) portfolio managers, analysts and finance specialists on average forecast the Fed to hike its policy rate by 29bp in 2016. The (money) market is pricing in about 20bp of hikes by end-year – or an 80% probability of a 25bp hike in 2016. In a Reuters poll conducted on 18 December analysts on average forecast 75bps of hikes. The 17 FOMC members’ median expectation, as per the December “dot chart” is for four 25bp hikes this year, which would take the policy rate to 1.375%. The weighted average of FOMC members’ expectation is for a slightly less hawkish 91bp of hikes.
2. 2 Figure 1: Someone is going to be wrong…and it’s more likely to be the Fed Figure 2: US Dollar trade weighted index likely to climb further Source: Reuters, FOMC Economic Projections (16 December 2015),www.olivierdesbarres.co.uk Source: Federal Reserve Bank of St Louis Therefore, by definition, US and global economic conditions are going to prove someone wrong which will in turn impact equity, bond and currency markets. I examine in turn the likelihood and implications of five main scenarios for Fed policy rates in 2016: i. 5 or more hikes ii. 3-4 hikes iii. 1-2 hikes iv. Rates on hold v. Rate cut(s) Scenario 1: Fed hikes five or more times in 2016 Probability: Very Low If the Fed hikes five or more times this year, this would mean a cycle of at least six hikes (including the December 2015 hike). In the past 25 years there have been four 12-13 month periods during which the Fed has hiked six or more times: February 1994-February 1995: Three 25bp hikes, three 50bp hikes and one 75bp hike = 300bps June 1999-May 2000: Five 25bp hikes, one 50bp hike = 175bps June 2004-May 2005: Eight 25bp hikes = 200bps June 2005-June 2006: Nine 25bp hikes = 225bp 0 20 40 60 80 100 120 Market Desbarres survey Reuters FOMC (weighted av.) FOMC (median) Expected US Federal Reserve rate hikes in 2016 (basis points) 80 90 100 110 120 130 140 Jan 95 Jan 02 Jan 09 Jan 16 Russia crisis great financial crisis US dollar trade weighted index (jan 97 = 100)
3. 3 Figure 3: US and global growth and CPI-inflation were considerably higher in previous hiking cycles Source: S&P/Case-Shiller U.S. National Home Price Index, Federal Reserve,FRED Data, IMF, Institute for Supply Management,U.S. Bureau of Labor Statistics, World Bank Note: 1. Consumer price index, % change year-on-year; 2. Core consumer price index, % change year-on-yea; 3. NOFP: Monthly change in total non-farm payrolls (seasonally adjusted); 4. US manufacturing Institute for Supply Management index; 5. TWI: US dollar broad trade weighted index (Jan 1997 = 100); 6. IMF methodology; 7. World Bank World Economic Monitor; 16th Dec 2015 +25bp to 0.375% GDP % qoq saar CPI % yoy (av.)1 Core CPI % yoy (av.)2 Unemployment rate, % (av.) NOFP, thousands (av.)3 ISM (av.)4 House price (% yoy) USD TWI level (av.)5 USD TWI (% change) Dow Jones (% change) GDP, % yoy6 CPI % yoy (av.)7 Core CPI- inflation % yoy (av.)7 3 months / quarter prior 2.0 0.1 1.9 5.0 235 49.6 4.8 120.5 1.9 6.1 3.0 1.3 2.2 15 Dec '15 - 27 Jan '16 0.7 0.6 2.1 5.0 292 48.2 5.3 124.0 1.9 -8.6 N.A 1.6 2.2 4th Feb 1994 +25bp to 3.25% GDP % qoq saar CPI % yoy (av.)1 Core CPI % yoy (av.)2 Unemployment rate, % (av.) NOFP, thousands (av.)3 ISM (av.)4 House price (% yoy) USD TWI level (av.)5 USD TWI (% change) Dow Jones (% change) GDP, % yoy6 CPI % yoy (av.)7 Core CPI- inflation % yoy (av.)7 3 months / quarter prior 5.4 2.8 3.1 6.6 282 55.1 2.1 N.A N.A 6.8 N.A N,A N.A Feb '94-Feb '95 (7 hikes, 300bp) 4.2 2.6 2.9 6.0 325 58.0 2.6 N.A N.A -0.6 N.A N.A N.A 30th June 1999 +25bp to 5.00% GDP % qoq saar CPI % yoy (av.)1 Core CPI % yoy (av.)2 Unemployment rate, % (av.) NOFP, thousands (av.)3 ISM (av.)4 House price (% yoy) USD TWI level (av.)5 USD TWI (% change) Dow Jones (% change) GDP, % yoy6 CPI % yoy (av.)7 Core CPI- inflation % yoy (av.)7 3 months / quarter prior 3.2 2.0 2.1 4.2 231 53.0 6.6 116.9 0.4 12.1 3.4 5.9 2.0 June '99-May '00 (6 hikes, 175bps) 4.2 2.7 2.1 4.1 278 55.8 8.7 116.4 2.5 -3.1 4.6 4.4 1.9 30th June 2004 +25bp to 1.25% GDP % qoq saar CPI % yoy (av.)1 Core CPI % yoy (av.)2 Unemployment rate, % (av.) NOFP, thousands (av.)3 ISM (av.)4 House price (% yoy) USD TWI level (av.)5 USD TWI (% change) Dow Jones (% change) GDP, % yoy6 CPI % yoy (av.)6 Core CPI- inflation % yoy (av.)7 3 months / quarter prior 2.3 2.3 1.7 5.7 296 60.9 12.0 115.8 1.5 0.8 5.9 2.8 2.1 June '04-May '05 (8 hikes, 200bps) 3.6 3.1 2.1 5.4 165 56.4 11.0 111.6 -4.9 -1.7 5.5 3.3 2.2 30th June 2005 +25bp to 3.25% GDP % qoq saar CPI % yoy (av.)1 Core CPI % yoy (av.)2 Unemployment rate, % (av.) NOFP, thousands (av.)3 ISM (av.)4 House price (% yoy) USD TWI level (av.)5 USD TWI (% change) Dow Jones (% change) GDP, % yoy6 CPI % yoy (av.)7 Core CPI- inflation % yoy (av.)7 3 months / quarter prior 4.3 3.1 2.2 5.2 225 52.7 14.3 110.8 1.7 -2.2 5.3 3.1 2.0 June '05-June '06 (9 hikes, 225bps) 3.2 3.6 2.1 4.9 212 54.8 8.8 110.5 -2.4 8.9 5.6 3.1 1.5 US data Global data US data Global data US data Global data US data Global data US data Global data
4. 4 Figure 3 summarizes key macro and market variables which likely influenced the Fed’s policy decisions before and during these four hiking cycles and compares them with the most recent data, cognisant that the Fed (and central banks generally) likely take into account a very broad spectrum of concurrent and forward looking indicators when setting interest rates. I highlight in red data points which are weak relative to the average five periods (and therefore not favourable to rate hikes), in yellow data points in line with the average and in green data points which are strong relative to the average (and thus favourable to rate hikes). US dollar strength or appreciation is typically disinflationary and therefore less conducive to rate hikes. The US unemployment rate and US and global core inflation in the past three months are broadly similar to the levels recorded in the previous four hiking cycles, at around 5%, 2% and 2% respectively (see Figure 3). The monthly increase in non-farm payrolls in September-November 2015, which averaged 235,000, was similar to the average increases in the run-up to the 1999 and 2005 hikes, while the December 2015 print of 292,000 is at the high end of the scale. US equities had a mixed performance prior the first hikes in 1994, 1999, 2004 and 2005, posting an average Dow Jones gain of 4.4% – versus a 6.1% rally in the run-up to the December 2015 hike. The Dow Jones has since dropped nearly 9%. But, importantly, US and in particular global GDP growth and headline inflation were significantly higher during these four hiking cycles, as was the US manufacturing ISM and house price inflation (see Figure 3). US GDP growth averaged 3.8%, almost three times as fast as the average 1.4% quarter-on-quarter seasonally adjusted annualised rate recorded in H2 2014. Decent US manufacturing supported overall growth during these previous hiking cycles, with the ISM in the mid-50s versus sub-50 in the past two months. US CPI-inflation averaged close to 3% in the run-up and during these previous hiking cycles, compared to only 0.1% yoy inflation in August-October 2015 and 0.6% yoy in November-December 2015. While the pace of US house price inflation has accelerated in recent months to around 5% yoy, it remains well below the average house price inflation of 11% yoy recorded in the run-up to the 1999, 2004 and 2005 hikes. Only in the run-up to the February 1994 hike was house price inflation more subdued at around 2% yoy. In the three months prior to 1999, 2004 and 2005 hikes, the US dollar trader weighted index (TWI) appreciated on average 1.2%. This disinflationary impetus was slightly weaker than the dollar TWI’s 1.9% appreciation in the three months to the 16th December 2015 hike and these currency gains came from much lower levels, particularly in 2005.
5. 5 Global GDP growth in 2004-2005 averaged 5.6%, almost twice as fast as in Q3 2015. It was only 3.4% yoy in Q1 1999 – not too dissimilar to current global growth of 3.0% – but then accelerated to 4.6% yoy in the following 4 quarters. This growth trajectory is unlikely this year with the IMF predicting growth closer to 3.4%, with history suggesting that this forecast may still be overly- optimistic (see What to expect in 2016 – same same, but worse, 19 January 2016). Impact In the unlikely event of the Fed hiking rates five or more times this year, which would imply a far more hawkish cycle than currently priced in, the US dollar’s path would likely depend on the global balance of growth and inflation. If the Fed is hiking because strong US and global growth are feeding off each other, it is possible to envisage a more subdued US dollar, with cheap EM currencies with strong trade/growth prospects outperforming. The US dollar would likely outperform if the US economy is strong and the Fed is hiking while the eurozone, Japan, Australia, Canada, China – facing weak growth – are forced to maintain loose monetary policy and Brazil, Russia and Turkey are unable to reflate their economies because of high inflation. But a decoupling of the US and global economy is an unlikely event, even if the US is a reasonably closed economy. Scenario 2: Fed hikes 3-4 times in 2016 Probability: Low The Fed hiking 3-4 times this year would imply roughly one hike every other meeting for the rest of the year and equate to a hiking cycle of 100-125bp (including the December 2015 hike) over a 12-13 month period (see Figure 4). This is broadly what FOMC members are forecasting on the premise of a stronger US economy gradually pushing headline inflation to 2%. There is no precedent in the past 25 years of the Fed hiking only 100-125bp over such a timeframe and the likelihood of the Fed breaking with history is low. Slowing real personal consumption growth, a struggling manufacturing sector, slowing global growth, very weak commodity prices and a strong dollar all point to US inflation remaining tepid this year, even if the headline number creeps up further. Furthermore, the very uncertain outlook for both developed and EM economies, the collapse in global equity markets and global risk aversion are, all things being equal, likely to curb the FOMC’s eagerness to further tighten monetary policy. Impact: Should the FOMC broadly stick to its script of four hikes this year, the impact on currencies, bonds and equities would depend on whether the market thinks such policy tightening is justified by underlying and forward-looking macro and market indicators and the likely impact on global growth and risk appetite.
6. 6 The US dollar may initially appreciate further as FX flows gravitate to higher-yielding US fixed income assets in a world of very low (or even negative) global rates. But if tighter US monetary starts to weigh more forcefully on the US economy and heightens the risk of global recession and deflation, the market’s appetite for US assets may wane and re-focus on economies where central banks are able and willing to support growth and on currencies which are historically cheap – the euro springs to mind. Figure 4: 2016 selected key dates Source: ECB, Federal Reserve,national statistics offices Scenario 3: Fed hikes once or twice in 2016 Probability: High Again, there is no precedent in the past 25 years of the Fed hiking only 50-75bp over a 12-13 month timeframe. Recent history shows that FOMC members have been overly-hawkish (see Figure 5). In the space of a year, FOMC members halved their average expected policy rate for end-2016 to 1.3%, in turn 01-Feb JP Morgan January global manuacturing PMI 05-Feb US January non-farm payrolls March China Politburo to unveil 5-year plan 06-Mar German municipal elections 10-Mar ECB policy meeting 13-Mar German state elections 16-Mar US FOMC meeting 13-Apr South Korea national elections 21-Apr ECB policy meeting 27-Apr US FOMC meeting 28-Apr US Q1 2016 GDP (advance estimate) 09-May Philippines Presidential elections 02-Jun OPEC meeting 02-Jun ECB policy meeting 15-Jun US FOMC meeting 21-Jul ECB policy meeting 27-Jul US FOMC meeting 29-Jul US Q2 2016 GDP (advance estimate) 08-Sep ECB policy meeting 21-Sep US FOMC meeting 20-Oct ECB policy meeting 28-Oct US Q3 2016 GDP (advance estimate) 02-Nov US FOMC meeting 08-Nov US Presidential elections 08-Dec ECB policy meeting 14-Dec US FOMC meeting
7. 7 partly due to the Fed systematically failing to accurately forecast inflation. In September 2014, the Fed was forecasting PCE inflation at 1.6-1.9% yoy in Q4 2015 – 1.5 percentage points above actual inflation or 1.5 times the standard deviation of PCE inflation using 20 years worth of monthly data. Figure 5: FOMC members have systematically revised down their Fed funds forecasts Source: Fed Board of Governors economic projections Market impact Market pricing of Fed rate hikes is very slim so FOMC members could slash their expected number of rate hikes this year without conceivably having a material impact on the dollar. There is a risk, however, that even if the Fed hikes broadly in line with market pricing – say in March or mid-year – the market will interpret this long time gap as the Fed’s tacit admission that it pulled the trigger too soon in December 2015, in turn pressuring the dollar near-term. Scenario 4: Fed keeps rates unchanged for rest of 2016 Probability: Low In the past 25 years, there is only one instance of the Fed delivering a one-off hike before staying on hold for a significant period of time (see Figure 6). 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 during the height of the Russian crisis. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 end-2015 end-2016 end-2017 Dec 2014 projections March 2015 projections June 2015 projections Sept 2015 projections Dec 2015 projections Weighted average of appropriate target level for the US federal funds rate (%) 2.54% 1.29%
8. 8 Figure 6: The lone hike Source: S&P/Case-Shiller U.S. National Home Price Index, Federal Reserve,FRED Data, IMF, Institute for Supply Management,U.S. Bureau of Labor Statistics, World Bank Note: 1. Consumer price index, % change year-on-year; 2. Core consumer price index, % change year-on-yea; 3. NOFP: Monthly change in total non-farm payrolls (seasonally adjusted); 4. US manufacturing Institute for Supply Management index; 5. TWI: US dollar broad trade weighted index (Jan 1997 = 100); 6. IMF methodology; 7. World Bank World Economic Monitor; But this point of comparison is clearly imperfect. The Fed’s policy rate was already very high at 5.25% so a hike to 5.50% could conceivably be justified as a tweak to policy, rather than the start of a new hiking cycle. This time round the starting point was very different – a 0.125% policy rate – and the Fed is clear that it foresees a series of albeit gradual rates hikes rather than a one-off adjustment. Impact If the Fed keeps rates on hold this year the market will likely interpret this as a tacit admission that the Fed pulled the trigger too soon in December 2015. This may near-term pressure the dollar, but if the market starts to price in rate cuts because the US and in particular global economy are weakening – an acute global risk aversion scenario – the dollar may perversely outperform. Scenario 5: Fed cuts rates in 2016 Probability: Low While probably still in a minority, those forecasting a US recession and Fed rate cuts this year are becoming increasingly vocal. I still think that it would take a significant US and/or global event for the Fed to reverse its recent hike. In the past 25 years there is no precedent for the Fed hiking rates once after a long hiatus only to cut rates a few months later and only once has the Fed cut rates when non-farm payrolls were robust. Despite monthly payrolls averaging 230,000 in June-August 1998, the Fed cut rates three times in succession in September-November 1998 – the orange shaded area in Figure 7. But this was at the peak of the Russia crisis and by end-1999 the Fed had fully reversed these cuts. Other rate cutting cycles have been associated with significant deteriorations in non-farm payrolls and on three 25th March 1997 +25bp to 5.50% GDP % qoq saar CPI % yoy (av.)1 Core CPI % yoy (av.)2 Unemployment rate, % (av.) NOFP, thousands (av.)3 ISM (av.)4 House price (% yoy) USD TWI level (av.)5 USD TWI (% change) Dow Jones (% change) GDP, % yoy6 CPI % yoy (av.)7 Core CPI- inflation % yoy (av.)7 3 months / quarter prior 4.3 3.1 2.5 5.3 285 54.0 2.6 101.2 3.8 5.0 4.8 5.1 2.6 April '97-July '98 (rates unchanged) 4.3 1.9 2.3 4.7 263 54.0 7.9 109.7 14.8 29.2 4.2 3.9 2.4 US data Global data
9. 9 out of four occasions with job losses – in mid-1990 to mid-1991, in 2001 and in 2008-2009 (the blue shaded areas). That is not to say that a reasonably robust US labour market precludes Fed rate cuts (as the 1998 crisis showed) but the hurdle for the Fed to cut rates is seemingly pretty high when domestic job creation is strong. Figure 7: Only once in 25 years has Fed cut rates when payrolls were robust Source: Federal Reserve,U.S. Bureau of Labor Statistics Impact At the risk of perhaps stating the obvious, such a policy u-turn would undoubtedly crush the Fed's already soft credibility and near-term may put the dollar under pressure. But if the Fed takes the unprecedented step of cutting rates so soon after a one-off hike, it is likely to be in the context of a severe deterioration in both the US and global outlook – put differently, it is difficult to conceive at present a scenario whereby the US economy is deteriorating rapidly but global growth and markets are recovering. In a scenario of extreme generalised risk aversion, history suggests that the dollar would either outperform as was the case during the 2008 great financial crisis or remain reasonably robust like in late 1998 (see Figure 2). -150 -125 -100 -75 -50 -25 0 25 50 75 -3000 -2500 -2000 -1500 -1000 -500 0 500 1000 1500 Jan-90 Aug-92 Mar-95 Oct-97 May-00 Dec-02 Jul-05 Feb-08 Sep-10 Apr-13 Nov-15 Change in Fed policy rate, bp (right scale) Change in US non-farm payrolls, 3-month rolling sum (thousands)
What if the Fed hikes, leaves rates on hold…or cuts? From olivierdesbarres.co.uk.
The US Federal Reserve kept interest rates ... US Fed leaves interest rates on hold. ... Fed Chair Janet Yellen was scheduled to hold a ...
Federal Reserve Leaves Rates On Hold; ... The average interest rate set by the Federal Reserve in the period from 1971 to 2015 (including, ...
The Federal Reserve left the target range for its federal funds rate unchanged at 0.25 percent to 0.5 ... Fed Leaves Funds Rate on Hold ...
... Home > US Fed leaves rates on hold. ... global events had diminished but not squashed chances of a rate hike in March. As part of the Fed's dual ...
Main Business Fed leaves interest rates on hold. 09.17 / 22:58 www.theage.com.au ...
Home > Market Wrap > Fed leaves rates on hold. Fed leaves rates on hold.
Federal Reserve leaves rates on hold. Wed 28 Oct 2015 18:00:07 GMT. Author: Adam Button ... Fed sees sold gains in household spending, business investment;
U.S. Fed leaves rates on hold. ... most economists thought the Fed would be raising rates by at least a quarter percentage point in August.
The US Federal Reserve kept interest rates ... rates on hold, preps for potential December ... hold a news conference on Wednesday. The Fed ...