Key takeaway – On September 16-17, the US Federal Reserve (Fed) will take a difficult decision: to hike rates for the first time in nearly a decade – or wait. It is a close call: US domestic economic conditions could justify tighter policy but remain fragile. While the GDP is growing, inflation is below the 2 percent target. Employment has improved, but wage growth and participation are low. Housing sector growth is buoyant but consumption is frail, with consumer confidence below expectations. Industrial production is sluggish and manufacturing is weak. Yet, non-manufacturing indicators are close to historical highs. In addition, the trade deficit narrowed and the USD has strengthened. At the same time, the global context looks increasingly weak: global growth is decelerating and leading indicators point to a slowdown. In most countries, retail sales are losing pace and – led by China’s drop – stock markets are on a downturn. Going forward, via trade and financial links the US economy will suffer. As a result, the Fed is likely to postpone the beginning of the tightening cycle to 2016. Yet, “the when” – what the markets are stressing about – matters little: what really matters is how long it will take to normalize monetary policy. The upcoming tightening cycle is likely to be well below historical norms and the markets should take notice. Markets’ reaction depends on expectations, and a long period of low interest rates is not priced in. In the long run, equities will suffer, bond yields rise and commodities fall.
On September 16-17 2015, the Fed will take a difficult decision. The Federal Open Market Committee (FOMC) will decide on the suitability of the first rate-hike since June 29, 2006. The decision is difficult because the US domestic conditions (i.e.: what matters the most) are in relatively good shape: the economy is growing, employment and housing are buoyant, non-manufacturing activity is strong, and the trade deficit has narrowed. In short, the state of the economy could justify tighter policy. Yet, the global context looks increasingly fragile: the European Union (EU) and Japan are stagnating, systemically important emerging markets (EMs) are decelerating (e.g.: China) or in recession (i.e.: Brazil and Russia), oil prices are volatile and financial markets increasingly jittery.
It is a close call, because of global links. A rate hike could exacerbate global concerns, add to the financial turmoil and – via trade and financial links – weaken the US economy. For example, periphery-to-core flows, triggered by risk aversion, could foster US dollar (USD) appreciation and EM currencies depreciation, lower commodity prices and inevitably stifle domestic growth and inflation. US leading indicators already hint at lower growth, industrial production is sluggish, inflation pressures are absent and wages low. Inevitably, Fed officials are closely monitoring a large set of data.
A. What do the data say? Here below we have tried to rebuild the key datasets.
1. The GDP is growing, due to stronger net exports, investments, and government spending.
- In 2Q2015, GDP grew at 3.7 percent year-on-year (y-o-y), revised up by the US Bureau of Economic Analysis from an initial estimate of 2.3 percent – a significant increase from 1Q2015 (0.6 percent), and above both its 2014 level and pre-crisis average.
- Personal consumption expenditures contributed 2.1 percent to growth, below its year-end 2014 reading. Net exports and government expenditure and investment reversed their negative contribution of 4Q2014, by adding 0.2 percent and 0.5 percent, respectively. Private domestic investment registered a contribution of 0.9 percent.
- Yet, for 3Q2015 leading indicators point to a slowdown and the Atlanta Fed forecasts GDP growth at 1.5 percent.
Since June 2009, GDP has averaged 2.2 percent per annum up until 2014, the slowest recovery in the past 70 years, more than 0.5 percent points weaker than in 2001-07 (the next slowest recovery).
Table 1. US – GDP growth accelerating across all categories, except consumption
- Impact on Fed’s decision: recent figures are positive, but leading indicators suggest that these levels of growth are unlikely be sustained in the near future. The economy does not appear strong enough to absorb an interest rate increase: the last time the Fed started a tightening cycle was in June 2004, when real GDP was growing at 3.0 percent, the pre-crisis average. Over the past four quarters real GDP growth has averaged 2.7 percent, in contrast to the two previous tightening cycles (3.4 percent in 1994 and 4.2 percent in 2004).
2. Inflation is below the 2 percent target, due to a strong USD, falling commodity prices, and slack in the US labor market. Upwards price pressures are likely to remain muted: the slowdown in commodity prices is not transitory (the Bloomberg Commodity Index is down -27.8 percent y-o-y, from -18.2 percent at the end of 2014), as it is largely due to a structural deceleration of China’s economy. The threat of outright deflation is receding, but inflation is declining (Table 2). Data for July show the following picture:
- The Personal Consumption Expenditures price index (PCE) stood at 0.3 percent y-o-y, down from 0.8 percent at the end of 2014, and below its 2.5 percent pre-crisis average. The Fed has a 2 percent target for PCE inflation.
- Core PCE grew by 1.2 percent y-o-y, down from 1.4 percent at the end of 2014, and below its 2.0 percent pre-crisis average.
- The Consumer Price Index (CPI) was unchanged at 0.2 percent, as in June, below both year-end levels (0.7 percent) and its pre-crisis average (2.9).
- Core CPI (excluding food and energy) grew by 1.8 percent, 0.2 percent points higher than at year-end 2014, but below its pre-crisis average (2.0 percent).
- In August, the Producer Price Index (PPI) remained unchanged from the previous month at 0 month-on-month (m-o-m), up from a -0.3 percent decrease in December 2014.
- With respect to inflation expectations, on September 10, 2015, the 10-year breakeven inflation rate was 1.59 percent, down from the year-end reading (1.68 percent) and its pre-crisis average (2.34 percent).
Table 2. US – Inflation is still subdued
- Impact on Fed’s decision: In the last four Fed tightening cycles, inflation was on the rise. The low inflation makes it difficult for the FOMC to argue that it has met its objective of price stability. Still, as monetary policy influences real activity with a substantial lag, the Fed might underweight inflation concerns in its decision making.
3. Labor market: employment indicators have improved, but wage growth and participation rate are still low. In 2015, a range of employment indicators have improved, although most remain below their pre-crisis average (Table 3). Yet, the labor market conditions index is sluggish and compensation indicators weak (Table 4). Data for August provide the following picture:
- The US economy added 173,000 jobs, down from 245,000 in July. Since May, the US has added an average of 235,000 jobs a month, up from a 193,000 pace between January and April.
- Unemployment rate declined to a seven-year low of 5.1 percent, down from 5.3 percent in July and within the 5.0-5.2 percent range that the Federal Reserve labels as “full employment”.
- U6, the broadest definition of joblessness (by including part-time workers for economic reasons, it better captures the slack), fell to a seven-year low of 10.3 per cent, from 10.4 per cent in July.
- Labor-force participation rate remained at 62.6 percent for the third straight month, the lowest level since September 1977 and below its pre-crisis average of 66.1 percent.
- In July, total non-farm job openings increased to a new high, reaching 5.8 million. The prior series high was 5.4 million in May 2015. The series began in December 2000.
- Little changed from June, in July, 2.7 million people quit their job. The number has been increasing since the end of the recession, but for the past 11 months has held between 2.7 and 2.8 million per month.
Table 3. US – Employment improving
- Impact on Fed’s decision: solid job growth and a diminished slack in the labour market increase the probability of a rate hike in September. In the past four Fed tightening cycles, employment growth was either flat or accelerating. The broader definition of unemployment U-6 is close to the level it displayed at the start of the two prior tightening cycles (11.4 percent in 1999 and 9.5 percent in 2004). Yet, labour market slack is higher than signalled by low unemployment numbers, as many individuals are not jobseekers and left the labor force.
- The US Fed Labor Market Conditions Index was more sluggish than payroll statistics: it rose by +2.1, as opposed to +1.1 in July, still far from the +7.2 at the end of December and an average of +4.6 since 2010.
- In 2Q2015, Employment Cost Index (ECI) data were weaker than expected, rising by +2.0 percent y-o-y (versus +2.6 percent in 1Q2015) and by +0.2 percent quarter-on-quarter (q-o-q), missing the expectations of +0.6 percent (+0.7 percent in 1Q2015), the smallest increase since records began in 1982.
- Also in 2Q2105, annual worker wage growth recorded +2.1 percent y-o-y, still below its pre-crisis average.
- Average hourly earnings rose +2.2 percent y-o-y, well below the Fed’s 3.5 target. Private payroll job growth slowed to 2.2 percent y-o-y from a cycle peak of 2.7 percent in February.
Table 4. US – Wages and participation rate still weak
4. Consumption is still fragile. Consumption is still weak (Table 5, but both income growth and firm employment trends suggest a future pick up in spending. Data for July show the following picture:
- Personal income rose by 0.4 percent m-o-m for the fourth straight month, still slightly below its pre-crisis average.
- Personal income y-o-y change grew at 4.3 percent, below the year-end level of 5.2, and its pre-crisis average of 5.6 percent.
- Personal spending grew at 0.3 percent m-o-m, back to positive as compared to December (-0.1), but still below its pre-crisis average.
- In y-o-y terms, personal spending rose by 3.5 percent, below year-end levels (4.0 percent) and its pre-crisis average of 5.7 percent.
- In August, retail and food services sales rose for the sixth consecutive month by 2.2 percent y-o-y, but still below year-end levels and pre-crisis average, at 3.3 and 5.1 respectively.
Table 5. US – Consumption still below pre-crisis averages
Impact on Fed’s decision: as purchases of goods bought on credit tend to fall when interest rates rise, the Fed will want to see strong growth in consumption before rising interest rates, to avoid a decline into negative consumption growth.
5. Consumer confidence is still missing expectations. Consumer confidence is well above pre-crisis levels, supported by low unemployment, lower gasoline prices and an improving housing market (Table 6). Data for August provide the following picture:
- The University of Michigan Consumer Sentiment index declined for the second consecutive month, from 93.1 in July to 91.9 in August (-1.2 points m-o-m), led by “current conditions” at 105.1 (-2.1 pts m-o-m from 107.2 in July) and “expectations” at 83.4 (-0.7 pts m-o-m from 84.1 in July).
- The current level of consumer confidence is up from a year ago (82.5) but below the cycle highs of 98.1 in January.
- The Conference Board’s Consumer Confidence Index hit a seven month high at 101.5 (+10.5 points), versus 91.0 in July and a cycle peak of 103.8 in Jan. The “labor market differential”, a variable closely correlated to the unemployment rate, was the most favorable since January 2008.
Table 6. US – Consumer confidence above pre-crisis averages, but below expectations
- Impact on Fed’s decision: in the past few months consumer confidence – which favorably impacts consumption – decreased since year-end, but it is higher than pre-crisis averages. If consumer confidence were to grow steadily over the coming months, consumer spending would increase and support the economic recovery. The Fed is more likely to raise rates once consumer confidence stabilizes.
6. In August, industrial production was sluggish. If coupled with declining consumer confidence (Table 6), this represents a downside risk to growth. Key data show the following picture:
- The Industrial production index grew by 0.9 percent y-o-y, below its pre-crisis average of 2.3 percent and year-end level of 4.6 percent.
- Capacity utilization, at 77.6 percent, did not improve from its pre-crisis or year-end level (79.0).
- The Manufacturing production index rose by 1.7 percent y-o-y, below year-end (4.2) and pre-crisis (2.9) levels.
- Manufacturing capacity utilization increased to 76.6 from 77.0 percent in July, below pre-crisis and year-end 2014 levels, at 77.1 percent.
Table 7. US – Industrial production still below year-end levels, capacity utilization is declining
Source: Federal Reserve Bank of St. Louis, 2015.
Impact on Fed’s decision: with industrial output not growing as fast as at the end of 2014, and declining from its July readings, a growth slowdown is possible. If growth were to decline, an interest rate hike could potentially trigger a recession.
7. In July, housing sector growth remained buoyant, with real estate indicators showing a strengthening housing-market activity. Yet, despite the buoyancy, new home sales are still below pre-crisis averages, raising questions about the sustainability of the housing-market recovery. Data for July show the following picture:
- Housing starts increased by 0.2 percent to 1,206,000 – the highest figure since October 2007 – from (an upwardly revised) 1,204,000 in June.
- Privately owned “one-family” housing starts increased by 58,000 (y-t-d) to 782,000.
- Existing home sales remained fairly robust at around 5.6 million units, the highest level since 2008, but still below the pre-crisis average of 6.31 million.
- At 507,000 units, new home sales are just above year-end figures, but only about half the pre-crisis average (1,078,000).
- The Pending Home Sales Index, a forward-looking indicator based on contract signings, rose by 0.5 percent, from 110.4 in June to 110.9, the third highest reading of 2015 behind April (111.6) and May (112.3). The index, now 8.2 points above December 2014 (102.5), has increased for the last 11 consecutive months.
- The NAHB Housing Market Index increased from 60 in July to 61 in August, the highest figure since November 2005, matching market expectations.
Table 8. US – Housing indicators improving significantly, if compared to year-end
Impact on Fed’s decision: while housing-market indicators show an improvement if compared to year-end 2014, they are still below pre-crisis averages. The weak trajectory of “new home sales” raises questions about the sustainability of the housing-market recovery, which could be hindered by an interest rate hike.
8. Manufacturing indicators are weak, but non-manufacturing ones are close to historical highs. The non-manufacturing sector is likely to drive growth in the coming periods, while a weak manufacturing Institute for Supply Management (ISM) index in August signals slower factory output. Data for August show the following picture:
- ISM Manufacturing Purchasing Managers Index (PMI) declined from 52.7 in July to 51.1 in August, the lowest level since May 2013, below both year-end (55.1) and pre-crisis (53.9) levels.
- ISM Non-Manufacturing Index (NMI) declined from 60.3 in July, the series all-time high, to 59.0 in August, the second highest reading since data are collected (January 2008), but it remains above year-end levels (53.1).
- ISM Non-manufacturing – Business Activity index declined to 63.9 from 64.0 in July, but still above both pre-crisis (59.0) and year-end (58.6) levels.
- ISM Manufacturing – New Orders Index declined to 51.7 from 56.5 in July, below the 57.8 of both year-end reading and pre-crisis average.
- ISM Non-manufacturing – New Orders Index declined to 63.4 from 63.8 in July, but it remains above its pre-crisis average (58.5) and year-end reading 2014 (59.2).
- ISM Manufacturing – Employment Index declined to 51.2 from 52.7 in July, below the 56.0 of December 2014, and the 51.8 pre-crisis average.
- ISM Non-manufacturing – Employment Index declined to 56.0 from 59.6 in July, but it remains above both pre-crisis average (53.3) and year-end figures (55.7).
Table 9. US – ISM: manufacturing indicators down, non-manufacturing indicators up
- Impact on Fed’s decision: the strong non-manufacturing data indicate that a robust services sector could support GDP growth in 3Q2015. At the same time, much weaker manufacturing data, perhaps on account of the strong USD, suggest a risk of further PMI declines. An interest rate hike by the Fed could negatively affect manufacturing indexes; only a resilient consumer growth would ease the risk of a sharp collapse in output growth.
9. The trade deficit narrowed in July, improving from it pre-crisis average, with both exports and imports contracting.
- The “goods and services trade balance” improved to -$41.9 billion (bn), from -$45.2bn in June.
- Both exports (-3.2 percent y-o-y, up from -3.6 percent in June) and imports (-3.1 percent y-o-y, down from -2.3 percent in May) are contracting, in contrast with the expansion of both exports and imports registered in December 2014 (1.1 percent and 3.8 percent, respectively), and if compared to pre-crisis averages (10.7 percent and 10.9 percent, respectively).
Table 10. US – Improved trade deficit, exports and imports are contracting
- Impact on Fed’s decision: an improving trade balance is supportive of a rate hike. When the trade balance improves, it supports real GDP growth and – by decreasing the supply of USDs in the global economy – it appreciates the USD.
10. Market indicators have improved in 2015, but they are still below pre-crisis levels.
- In July, the Effective Federal Funds Rate was 0.13 percent, slightly above its year-end value (0.12) but below iys pre-crisis average (3.13).
- The 2yr and 10yr Treasury Constant Maturity Rates are at 0.75 percent and 2.23 percent, respectively, both slightly above year-end values (0.67 and 2.17) but significantly below pre-crisis averages (3.41 and 4.40).
- The BofA Merrill Lynch US Corporate 7-10 Year Effective Yield stood at 3.98 percent, up from the 3.68 of year-end 2014 but below its pre-crisis average (5.45).
- In June, the Case-Shiller 20-City Composite Home Price Index increased by 5.0 percent y-o-y, above year-end levels (4.4).
Table 11. Improved market indicators in 2015, still below pre-crisis levels
- Impact on Fed’s decision: all indicators show room for a hike. The Federal funds rate – near zero since late 2008 – needs to return to be one of the Fed’s primary instruments for steering the economy. Yields on 2-year and 10-year treasuries – benchmarks for the rates paid on loans – reflect longer-term expectations for inflation and Fed policy. Data show that the yield curve is normalizing. Finally, household wealth indicators (Case-Shiller Home Price Index, and stock markets – see Table 15) are relatively stable and influence consumer spending.
11. The USD has strengthened due to fundamentals – as the US economy is growing faster than peers – and to market sentiment: financial markets jitteriness and the possibility that the Fed might hike in September have created periphery-to-core flows. The Trade-Weighted US Dollar Index (Figure 1) has improved 18.1% year-to-date (y-t-d), and is currently at its highest level since August 2003.
Figure 1. The USD has strengthened
Note: The trade-weighted US Dollar index is a weighted average of the foreign exchange values of the U.S. dollar against the currencies of a large group of major U.S. trading partners. The index weights, which change over time, are derived from U.S. export shares and from U.S. and foreign import shares. The Index is calculated against a basket of 27 currencies. The currencies with maximum trade weights (in use since October, 2014) include- CNY(21.29%), EUR(16.38%), CAD(12.66%), MXN (11.87%), JPY(6.90%), KRW (3.87%) and GBP(3.35%). January 1997 = 100. 10-year Average and standard deviation calculated for the period: 12/09/2005 – 11/09/2015.
- Impact on the Fed’s decision: a stronger USD could damage the economy. A rate hike would strengthen the USD further, reducing exports and hampering growth.
12. Global growth is slowing. Global growth is projected at 3.1 percent in 2015, lower than 3.4 percent in 2014. Developed markets (DMs) are stagnating and EMs slowing down. Over the years, many countries reduced spending and investments, depressing aggregate demand. As a result, export – especially in China and Germany – stalled, and commodity prices fell, creating deflation risks. Concerns about China’s growth caused financial market jitteriness (Table 15). Beijing joined the currency war by devaluating the yuan by 2.9 percent between August 11-20, 2015. In 2016, growth is expected at 3.2 percent, but a global recession (i.e.: global growth below 3.0) is not unlikely.
Table 12. Global – Only US GDP above pre-crisis average
Leading indicators point to a slowdown. With the exception of the EZ, PMI is decreasing in all main economies, retail sales are slowing their growth in most countries, and the stock markets are growing at a slower pace since the summer.
Table 13. Global – Only EZ PMI is recovering, the rest is deteriorating
Source: Trading Economics, 2015
Table 14. Global – Retail sales growing at slower pace
Source: Trading Economics, 2015.
Table 15. Global – Stock markets on a downturn after China’s stock market drop
Source: Bloomberg, 2015.
B. When will the Fed rise interest rates? Likely next year … As above indicated, despite improvements in the labor market, inflation remains below the target of 2 percent. Recent financial markets volatility has added uncertainty to a challenging global environment. The risk of making a decision that might have to be reverted in the near future will force a cautious approach: not to lose credibility, the Fed might delay rate hikes to 2016. During the year, the FOMC holds eight scheduled meetings, and others as needed. According to the July 29 FOMC statement, policy makers want to see “some further improvement” in the labor market before they hike, and the minutes show that the conditions for tightening are approaching, but not yet been achieved. For the rest of 2015, scheduled meetings are on September 16-17, October 27-28, and December 15-16. The September and December meetings – by including a press conference by the Chair and the release of the Summary of Economic Projections – are the most likely for a policy announcement.
- Scenario 1 (20 percent): a 25 basis points hike on September 16-17. In this scenario, the FOMC will focus on the fact that labor market data show broad-based employment gains. The 235,000 average rise in payroll jobs since May will be taken as another sign of labor market strength. Currently, the Fed futures markets discount a 28 percent probability of a hike, and FedWatch assigns a 25 percent probability to this scenario.
- Scenario 2 (30 percent): a 25 basis points hike in December. In this scenario, the FOMC will highlight that wages remain soft, and – in the language of policy makers – qualify as ‘further improvement’. Tepid inflation, low participation rate, the Atlanta Fed’s real GDP estimate for 3Q2015 at just 1.5 percent, and uncertainty about global developments would justify opting for a delay.
- Scenario 3 (50 percent): no hike this year. In this scenario, the FOMC will reassess its assessment of the external environment. Since the summer, EM capital outflows accelerated. EM central banks intervened – by selling DM government bonds they hold as foreign exchange reserves – to defend exchange rates. This “quantitative tightening” might undermine global asset prices and delay the hike. Additionally, an already strong USD is hurting American companies with operations abroad, which negatively impacts US trade.
… however, what matters is the end, not the beginning of the tightening cycle. The key question is not when the Fed will start the tightening cycle, but and how long it will take to normalize monetary policy, and by how much. In other words, what matters is the pace of the successive hikes, the duration of the tightening cycle and the end level of the interest rate (Table 17).
Past tightening cycles cannot be a reference … Since 1965, the Fed has embarked in 15 policy tightening cycles. On average, the duration was 14 months and the increase in interest rates 411 basis points. The median duration was 12 months, and the median increase in interest rates 300 basis points (Table 16).
Table 16. US Fed – Tightening cycles since December 1965
… as the upcoming tightening cycle is likely to be well below historical norms. Normalization of monetary policy implies a return to potential growth and full employment. It will not happen soon. Given existing and forecasted macroeconomic conditions, the process will be unusually gradual, especially if compared to the last tightening cycles (Table 16). Each cycle depends on macroeconomic and financial-market idiosyncrasies: this time the Fed has to manage a transition from non-traditional to traditional monetary policy in a fragile context. As a consequence, the tightening pace is likely to be drastically slower, the average hike size more modest, the duration significantly longer, and the peak rate lower than in previous cycles (Table 17). Even Fed policymakers expect a gradual cycle, as suggested in June’s “Summary of Economic Projections”. The markets should take notice.
Table 17. US Fed – policy normalization scenarios: the slowest, longest tightening cycle ever
Source: Authors’ elaboration, 2015.
Note: * one every other FOMC meeting (there are 8 meetings per year).
C. Markets: too much focus on the first hike, a long period of low interest rates is not priced in
Historically, markets’ reaction largely depends on expectations … Normally, a highly anticipated decision has little impact on the markets (Table 18).
Table 18. US Markets – Reaction during the last 15 Fed tightening cycles
Source: Deutsche Bank, 2015.
… but the fist hike is not the issue! The markets should expect the slowest tightening pace in history. The markets are focusing on the timing of the first hike focus, when the real issues are pace and duration of the tightening cycle. After the start of the 1994 tightening cycle, financial markets were caught by surprise with regard to the timing and the scale of rate hikes, and responded negatively (Table 18).
- Equities: neutral in the short run, to suffer in the long run. In theory, stocks are expected to react negatively to an increase in interest rates: corporate debt-servicing increases, a decline in consumption hits revenues and the estimated future cash flows is reduced. A stronger USD negatively affects multinationals, exporters, and domestic products that compete with imports. However, data for the last Fed tightening cycles show that higher rates are not always bad news. In ten of the 15 most recent cycles, the stock market has risen in the year following the first rate increase. In other words, there is no direct correlation between the start of a tightening cycle and a fall in equities (Table 18). Finally, the Fed’s first has been widely anticipated and it would not come as a surprise. However, the abnormal length of policy normalization is not priced in and will likely hit equity markets in the long run.
- Bond yields to rise. Higher target rates will raise bond yields, reducing the prices of bonds. During all-but-one of the last 15 tightening cycles, treasury-bond yields rose (Table 18). In average, long-duration assets suffer the most, but rate rises eventually attract capital into Treasury bonds. Overall, fixed income stand to benefit from the global concerns about global growth and inflation which weigh on all asset classes. The risk is a repeat of 2013 “taper tantrum”, when panic caused money to abandon of the bond market, resulting in a dramatic increase in bond yields.
- Commodities to fall. Most commodities are priced in USD: a further strengthening of the USD will put downward pressure on prices, as commodities become more expensive for holders of other currencies. Gold – as an investment that hedges against inflation – will see further drops if a rise in rates is faster than market has anticipated: prices of gold are already down over 6 percent year-to-date. Conversely, if rate hikes lead to slower global growth, gold prices would increase driven by the perception of “safe haven asset”.