June 2014 Report Series

Inflation, Employment and the Fed’s Calculations: The Outlook for US and Saudi Interest Rates Executive Summary

Economics Department Samba Financial Group P.O. Box 833, Riyadh 11241 Saudi Arabia [email protected] +44 207659-8200 (London) This and other publications can be Downloaded from www.samba.com



With the withdrawal of US quantitative easing now well under way, attention is shifting towards the timing of the first US policy rate hike. Given that the Saudi riyal is set to remain pegged to the US dollar, any rate hike will feed directly through to Saudi rates, meaning higher borrowing costs for Saudi consumers and investors.



The Federal Reserve’s “dual mandate” of full employment and price stability makes it difficult to predict when rates might rise, especially as the Great Recession appears to have ruptured the relationship between inflation and employment. The overall rate of unemployment has fallen quite quickly over the past couple of years and is rapidly closing on the Fed’s target rate. However, inflation has remained remarkably subdued, leading some—including the Fed Chair, Janet Yellen—to wonder whether the labour market is carrying more economic slack than might be apparent from the headline unemployment figure. As such, Ms Yellen has emphasised that the Fed will consider various labour market indicators, including the number of long-term unemployed and the participation rate, when determining the timing of any rate hike. She also appears to consider deflation, rather than inflation, more of a threat to the economy.



In this paper we will consider whether Ms Yellen’s fears are justified: is there excessive slack in the US economy? There may be large numbers of long-term unemployed, but do they have much influence on wages and prices? Can deflation really be considered a threat, particularly when inflation expectations appear so well anchored? Put another way, is the Fed’s accommodative stance storing up trouble for the future? Ultimately, we want to know how far Ms Yellen’s dovish stance is shared by other members of the Fed’s Open Market Committee and what this means for the timing of any rate hike.

June 2014

Introduction: the shadow of deflation

Fed Chair Janet Yellen points to excess slack in the US economy—and the dangers of deflation

In a recent speech in New York, the Federal Reserve’s Chair, Janet Yellen, outlined her general view of the US economy and the questions that the Fed would be considering when deciding about the path of US policy rates. Consistent with the Fed’s dual mandate to foster “maximum employment in a context of price stability” Ms Yellen focused both on labour market indicators and price pressures. The Fed Chair observed that inflation has remained remarkably subdued even as the jobs market has recovered. Ms Yellen suggested that this might indicate that the labour market is carrying more slack than currently recognised— slack that is not captured by the current headline rate of unemployment (6.3 percent and falling). In fact, Ms Yellen indicated that deflation might become a problem for the US economy. In a thinly veiled reference to Japan she noted that “once it starts, deflation can become entrenched and associated with prolonged periods of very weak economic performance”. In this paper we will consider whether the US economy is indeed carrying excessive slack, and what that might mean for price and jobs growth. Ultimately, we want to know how the debate about slack might affect US policy rates. With the SAR/USD exchange rate peg almost certain to remain in place, any US policy rate changes are likely to be replicated by the Saudi monetary authorities, meaning higher borrowing costs for Saudi businesses and consumers.

What is economic slack and how much is the US economy carrying? Our first task is to define economic slack. In her speech, Ms Yellen defined slack as “the shortfall in employment relative to its mandate-consistent level”. This translates as the difference between the “natural” level of unemployment and the current level of unemployment. Technically, the natural rate is known as the non-accelerating inflation rate of unemployment, or NAIRU— that is, the level of structural unemployment consistent with no inflation pressure. It could also be termed the Fed’s target rate of unemployment. For the FOMC members, the natural rate of unemployment ranges from 5.2-5.6 percent of the labour force. By comparing this rate (or range) with the current rate of unemployment, one might surmise that there is still some slack in the economy, but it is a gap that is shrinking quickly. The Great Recession saw the overall unemployment rate soar to 10 percent in 2009, but in recent years the rate has come down rapidly. By June 2014 the rate was 6.3 percent—not far from the natural rate.

2

June 2014

Inflation has remained subdued in the past couple of years despite a 3 percentage point decline in unemployment

What puzzles Ms Yellen, and others, is that the corresponding upturn in inflation has been so muted. Consumer price inflation has yet to make a convincing move above 2 percent and is still well below its 2011 peak of 4 percent. The Fed’s preferred inflation measure is personal consumption expenditure (PCE) inflation. The PCE core rate (that is excluding volatile food and energy prices) has recently turned up a bit, but it remains well below the 2 percent registered in 2012 and some way off the precrisis average of around 2.2 percent. Price growth has therefore weakened during a period when unemployment has eased by almost three percentage points.

Janet Yellen’s View: There is excessive slack in the economy and it represents a threat This puzzle has preoccupied Janet Yellen. She thinks that weak inflation in a context of strong jobs growth suggests that there is in fact a lot more slack in the jobs market than most people think. She focuses on two elements of the labour market that appear to support her suspicion: long-term unemployment and the participation rate. The ratio of long-term unemployed (defined as those out of work for 27 weeks or more) remains persistently high at just over 35 percent of the jobless total; the pre-crisis ratio was well below 20. In her New York speech (and in other fora) Ms Yellen said “I believe that long-term unemployment might fall appreciably if economic conditions were stronger”. There is a similar story with the participation rate—not surprising since the two are closely related. The participation rate measures those working or looking for work as a proportion of the total population. Having been stable, at around 66-67 percent for much of the preceding two decades, the participation rate plunged with the onset of the Great Recession, and has shown little sign of stabilising since; in June 2014 the rate was at 62.8 percent. Much of this decline represents long-term unemployed who have “given up”, but again, Ms Yellen believes that the low participation rate might be pointing to slack that is not captured by the headline unemployment rate. That is, more people would re-enter the labour force if they thought that the economy were picking up again (indeed, the unemployment rate has been declining quite quickly partly because the participation rate has declined). If there is significant slack in the US economy, why has it persisted? Why for example has the long term unemployment rate remained so stubbornly high and why has the participation rate continued to decline? In Ms Yellen’s view, most of the blame

3

June 2014 can be attributed to the severity of the Great Recession, which pushed many US workers into long periods of unemployment—a process so dispiriting that many workers simply gave up and stopped looking for work (hence the declining participation rate). For Ms Yellen, the fact that there are so many “marginally attached” or long-term unemployed workers is bad in itself. But it also carries a systemic threat: that of deflation, which begets a spiral of declining prices, weakening demand and soaring debt burdens. Ms Yellen’s concern is that the surfeit of workers will keep downward pressure on wages to the extent that inflation expectations will eventually become unhinged. Inflationary expectations are difficult to budge (see below) but if they are, thinks Ms Yellen, then the US could be quickly dragged down into a deflationary spiral (a threat that appears to be taking shape in the Eurozone).

An Alternative View: there is not much slack and loose monetary policy is storing up trouble But are Ms Yellen’s fears justified? Some believe that there is not, in fact, very much slack in the US economy and an unnecessarily loose policy stance might be storing up much higher inflation in the future. With short-term unemployment almost back to normal, that would mean wages are about to accelerate and interest rates would have to go up sharply and soon. Citi, for one, believes that inflation persisted during the Great Recession precisely because there was not much slack in the economy1. But why would there be little slack during such a hard-hitting recession? Citi argues that the recession caused a fundamental rupture in the relationship between inflation and unemployment. That is, the recession was so severe that jobs were “permanently” lost and skills became redundant: hence the large numbers of long-term unemployed. According to this argument, people who are structurally unemployed have less downward influence on wages since they are, to all intents and purposes, unemployable (unless they retrain). Exacerbating this situation is the retirement of the Baby Boomer generation, which has contributed to the decline in the participation rate. Thus, for Citi, although unemployment was high, inflation persisted during the Great Recession because a large swathe of the unemployed were effectively no longer in the labour market, and thus could not put downward pressure on wage growth. Citi illustrate this rupture by means of the Phillips Curve. The Philips Curve attempts to capture the relationship between inflation, employment and inflationary expectations, and is used to predict

4

June 2014 inflation. The basic formula for the Phillips Curve is:

where is inflation expectations and represents the level of unemployment. is a parameter commonly referred to as the slope of the Phillips Curve, which captures the strength of the relationship between inflation and cyclical unemployment. Thus, a steep Phillips Curve suggests a tight relationship between inflation and cyclical unemployment, while a flatter curve suggests that the relationship is much looser. Up until the Great Recession, the Phillips Curve was pretty good at predicting inflation (especially under Citi’s augmented model, which includes supply-side innovations such as changing productivity trends and commodity price movements). However, with the onset of the GR, the relationship appears to break down, with the Phillips Curve anticipating deflation, when in fact there was none (see chart). Citi therefore opts to use other labour market indicators that might be more insightful than the overall unemployment rate. These include: the short term unemployment rate, the level of non-farm payroll employment, initial claims for unemployment insurance, and the number of job openings in the labour market. Citi plugs these other labour data into their augmented Phillips Curve and finds that they all imply much lower unemployment than that provided by the overall unemployment rate. In other words, the unemployment rate is “wrong” and there is far less slack in the economy than implied by this overall rate. The implications of this and other studies are that the economy is much closer to over-heating than many think. Other research houses offer a similar line, pointing to the recent pickup in: inflation (albeit from a low base), consumer loans, companies’ compensation plans, and commercial rents as evidence.

What does the Fed think? How has this debate influenced the Fed? First, it should be noted that Ms Yellen’s concerns about long-term unemployment, a falling participation rate, deflation, etc, are essentially risks to a benign Fed outlook. The FOMC’s “current outlook” is that the economy will continue to recover and that unemployment will fall into the range of 5.2-5.6 percent (the “natural” range) by the end of 2016. By this point, inflation should have moved into a 1.7-2 percent range (on the PCE measure). These targets do not have to be hit before the first rate rise is introduced; rather, as Ms Yellen makes clear, the FOMC will “assess progress, both realised

5

June 2014

Many Fed members appear to agree with Ms Yellen that there is slack in the economy

and expected” towards these goals when considering the timing of “liftoff” as it is known. But the Fed’s forecast is only a baseline. It involves a considerable degree of uncertainty and is liable to change. Fed policy statements give hints of the balance of opinion and how this might be evolving. Notably, the April FOMC minutes record that: “a number of participants expressed scepticism about recent studies suggesting that long term unemployment provides less downward pressure on wage and price inflation than short term unemployment does.” This appears to challenge directly the research of Citi and others. The minutes instead make mention of “other research findings that both short- and long-term unemployment rates exert pressure on wages” in reference to a working paper by Fed staffer Michael Kiley2. Mr Kiley uses more granular data—using regional sources—to demonstrate that the short-term and long-term unemployed exert equal downward pressure on price inflation2.

The Fed appears committed to looking at longterm unemployment, the participation rate, etc, when deciding about the next interest rate rise

Thus, one might surmise that a number of participants on the FOMC—probably a majority and including the Chair—believe that there is still considerable slack in the economy, and that slack is manifest in the high number of long-term unemployed, the low participation rate, and the weak inflation readings. As the April minutes make clear, in deciding how long to keep the Fed Funds Target Rate at the current 0-0.25%, the Fed will: “take into account a wide range of information, including measures of labour market conditions, [and] indicators of inflation pressures and inflation expectations”. Taking the long-term unemployed, the number has been declining at a monthly pace of about 7 percent for the past two years—not bad, but at that pace the overall figure would only reach pre-recession levels in about five years from now. Meanwhile, the participation rate would probably take about six years to regain its pre-recession levels, assuming that it stops falling and begins to reverse in the next couple of months (far from guaranteed). Does this mean that the Fed is going to delay raising the policy rate for five or six years? Most probably not, but it does indicate that the labour market still has plenty of “healing” to do. Another way of looking at this is through the Fed’s own analysis of the fortunes of the labour market during recessions and expansions.

6

June 2014 The Fed has devised a Labour Market Conditions Index (LMCI) which is a dynamic factor model of various labour market indicators that the Fed thinks summarise overall labour market conditions fairly well. One of the elements of the LMCI is the participation rate, which can be taken to correlate strongly with the long term unemployment rate (long term unemployment is not included as a separate element on the LMCI3). Tracking the change in the LMCI during contractions and expansions since 1980, the Fed finds that the index sheds an average of 219 points during a recession and gains about 207 points during an expansion—a difference of 12 points. In the current expansion, which began in July 2009, 290 points have so far been added to the LMCI—a decent performance when judged against the average. However, the LMCI shed 370 points during the Great Recession, which means that the index needs to regain a further 68 points (370 minus 290 minus 12) to get back to par. Since 2012 the LMCI’s monthly gain has averaged about 4 (this also happens to be the long-run average for LMCI gains during expansions). At this rate, the recovery in the labour market will take a further 17 months, i.e. until September 2015. Given that the LMCI includes the participation rate, one can project that the long-term unemployment rate will come down over the next seventeen months. Given also that this is a key element of slack, one might also expect the threat of deflation to dissipate.

Could inflation or deflation derail the interest rate hiking cycle? Of course, it is unwise to take too mechanical an approach to the Fed’s “reaction function”. There is no guarantee that the labour market will continue to heal at a steady pace. Unexpected events could blow the economy off course and bring the prospect of deflation back into play. Alternatively, one might take the Citi view that the Fed’s unnaturally accommodative policy stance means that inflation is far more of a threat than many realise and that inflation expectations are likely to become unhinged as inflation picks up rapidly. In fact, the evidence suggests that it is very difficult to dislodge inflation expectations. Analysis from the IMF4 shows that inflation expectations have become less responsive to actual inflation in recent decades and far more anchored around central bank targets—despite quite large deviations between targets and reality in some countries.

7

June 2014 Moreover, in Ireland, Spain and the UK unemployment actually fell below the NAIRU during the 2000s. And Spain and Ireland also had inappropriately loose monetary policies given their membership of EMU. Yet inflation and inflation expectations remained remarkably stable in all three countries5. There is less evidence with regard to deflation (since it is a much rarer phenomenon) but it does seem unlikely that inflationary expectations would move into deflationary territory while the economy is growing. Even at the height of the financial crisis in December 2008, when asset prices, consumer prices, and output were collapsing, inflationary expectations remained close to 2 percent.

Bottom Line: consensus seems about right The Fed’s dual mandate makes predicting the first interest rate tricky, particularly as the Great Recession appears to have disrupted the relationship between inflation and unemployment, while measuring the strength of the labour market has itself become more difficult. For all that, we tend to agree with Ms Yellen that the elevated rate of long term unemployment points to at least some slack in the US economy. However, unlike Ms Yellen, we do not see deflation as a threat and expect that this slack will be removed as the economy picks up pace over the next year or so, as the change in the LMCI would appear to indicate. With the LMCI heading back to par and the threat of deflation receding, the consensus forecast of the first hike in the Fed Funds rate occurring in the third quarter of 2015 appears about right to us. We think that the initial rise will be of 50 basis points, with no further increases in 2015 as the Fed assesses the impact of its hike. Assuming the hike does not lead to major economic and financial disruption then we would expect the tightening cycle to continue in 2016 with a further two 25-basis-point rises in the first half, and a further three in the second half (though more likely as one 50 bps hike and one 25 bps hike). This would leave the Fed Funds Target Rate at 2 percent by end-2016. There is plenty of risk to this forecast. The US economy is struggling to gain “escape velocity”: even now, the median US household is still almost 8 percent poorer in real terms than it was in 2008. With the middle still squeezed, and housing showing signs of going into reverse, there is clearly a risk that consumption could grind lower in the next year or so. As such, significant slack might persist for longer than many expect, meaning that any increase in policy rates is delayed.

8

June 2014

What does this mean for Saudi interest rates? Assuming that these risks do not materialise, the analysis and forecast presented in our April 2013 paper on Saudi interest rates still largely holds6. In this, we noted that the Saudi riyal’s peg to the US dollar is expected to remain intact, and therefore any rise in the US Fed Funds rate will be replicated (if not immediately) by SAMA in its reverse repo rate. This in turn would be transmitted to SAIBOR, the main Saudi interbank rate. The other main influence on Saudi rates is domestic demand for credit. We expect this to continue growing over the next few years, but at a softening rate as the economy cools in line with weakening oil prices. Thus, we think that SAIBOR will continue to drift lower this year and next, before the 50 bps increase in the Fed Funds rate is delivered in the third quarter of 2015. This would mean that SAIBOR ends 2015 at around 1.2 percent. The rate of credit growth is set to continue cooling in 2016 (partly in response to higher rates) but with a further 125 bps of policy rate rises feeding through, we expect SAIBOR to finish 2016 at just over 2 percent.

Higher interest rates are in the offing for KSA, but the peak should be lower than in the 2006-08 period.

Note that the US tightening cycle in the coming years is expected to be shallower than previous cycles. Consensus projects the peak of the forthcoming cycle to be reached in the first half of 2018, when the Fed Funds rate is expected to be 3.75 percent, i.e. an increase of 3.5 percentage points in two and a half years. This compares with the 2004-2006 cycle when rates were raised by 4.25 percentage points in two years. Moreover, at 3.75 percent, the peak is considerably below previous highs, and is in fact well below the average level of the Fed Funds target rate for 19902008 (4.25 percent). Similarly, the projection for SAIBOR, which we see peaking at a similar level to the Fed Funds in 2018, would still be below the average for the past decade of about 4 percent; indeed, for much of 2006-08, SAIBOR was at or near 5 percent. In short, although higher interest rates are clearly in the offing, this is far from “uncharted waters” for the Saudi economy. The Saudi economy is used to significantly higher rates, and should be able to cope without undue stress.

9

June 2014

1

Missing Inflation? Explaining the Behaviour of U.S Consumer Prices, Benjamin R Mandel, Joe Seydl, Citi Research, May 2014. 2

Michael T. Kiley: An Evaluation of the Inflationary Pressure Associated with Short- and Long-term Unemployment. Office of Financial Stability and Division of Research and Statistics, FRB, Washington, D.C., March 2014 3

Details of the data, model, and estimation will be presented in a forthcoming FEDS working paper. See Assessing the Change in Labor Market Conditions, Chung et al, FEDS Notes, May 2014. 4

The Dog that Didn’t Bark: Has Inflation Been Muzzled or was it Just Sleeping? IMF World Economic Outlook, April 2013. 5

That is not to say that all forms of inflation were subdued. Asset price inflation—particularly in housing—was a feature of all three countries. As such, low consumer price inflation does not necessarily mean a lack of economic imbalances. 6

Saudi Arabia: Interest www.samba.com.

Rate

Outlook,

2013-15,

April

2013,

10

June 2014 James Reeve Deputy Chief Economist [email protected] Andrew Gilmour Deputy Chief Economist [email protected] Thomas Simmons Economist [email protected]

Disclaimer This publication is based on information generally available to the public from sources believed to be reliable and up to date at the time of publication. However, SAMBA is unable to accept any liability whatsoever for the accuracy or completeness of its contents or for the consequences of any reliance which may be place upon the information it contains. Additionally, the information and opinions contained herein: 1.

2. 3.

Are not intended to be a complete or comprehensive study or to provide advice and should not be treated as a substitute for specific advice and due diligence concerning individual situations; Are not intended to constitute any solicitation to buy or sell any instrument or engage in any trading strategy; and/or Are not intended to constitute a guarantee of future performance.

Accordingly, no representation or warranty is made or implied, in fact or in law, including but not limited to the implied warranties of merchantability and fitness for a particular purpose notwithstanding the form (e.g., contract, negligence or otherwise), in which any legal or equitable action may be brought against SAMBA. Samba Financial Group P.O. Box 833, Riyadh 11421 Saudi Arabia

11