- The gradual normalisation process being conducted by the US Federal Reserve (Fed) is well underway: the central bank has raised interest rates six times since December 2015 and began reducing its bloated balance sheet in October 2017.
- Judging by its latest projections, the Fed is about halfway through its current hiking cycle and it foresees two or three additional rate hikes in both 2019 and 2020. During this time, the Fed expects quantitative tightening (QT) to run in the background without disrupting markets.
- Uncertainties about the Fed’s estimate of equilibrium rates, or r*, and future economic developments mean that its projections for the policy rate should be taken with a pinch of salt.
- Although the Fed will probably deliver two or three additional rate hikes this year, a persistently low r* and economic disruptions might lead to slower and limited interest rate hikes in the coming years.
- Another significant source of tightening is balance sheet normalisation. If the Fed’s QT proceeds as planned, its balance sheet will shrink by about $1tn by the end of 2019. According to our estimates, that is equivalent to an additional 130bps of tightening.
- Several potential challenges and risks could impact the Fed’s course of action, including significant fiscal stimulus, the evolution of the US yield curve, a potential overshoot of inflation and, most importantly, the possible escalation of protectionist threats. On balance, we believe normalisation is likely to be slow rather than fast.
- Recent personnel changes at the Fed are unlikely to disrupt the policy course in the short term, but they might affect its broader framework in the medium to long term. However, some new Federal Open Market Committee (FOMC) members have already advocated a shift towards a more rules-based approach to monetary policy and criticised unconventional policies.
“Participants generally agreed with the assessment that continuing to raise the target range for the federal funds rate gradually would likely be appropriate if the economy evolves about as expected. These participants commented that this gradual approach was most likely to be conducive to maintaining strong labour market conditions and achieving the symmetric 2% inflation objective on a sustained basis without resulting in conditions that would eventually require an abrupt policy tightening1."
The message from the minutes of last month’s FOMC meeting is clear: the Fed will continue to gradually raise official interest rates in the short term, a strategy that has proved successful thus far.
Today, the Fed’s normalisation process is well underway. It involves the two primary levers of the current monetary policy framework: the policy rate and balance sheet.
Since the Fed began the current hiking cycle in December 2015, it has raised the policy rate six times, from the zero-lower bound, which prevailed in the years following the financial crisis, to the current range of 1.5% to 1.75%. Meanwhile, the Fed introduced Quantitative Tightening (QT) – the reversal of Quantitative Easing (QE) – in October 2017, which follows a plan to gradually reduce the size of its hefty balance sheet. The bank has embarked on the path to QT in a passive manner, simply tapering the reinvestment of maturing Treasuries and Mortgage-Backed Securities held on its balance sheet.
In the coming years, the Fed expects to maintain its gradual normalisation course, supported by a constructive outlook for the US labour market and a gradual convergence of inflation to the 2% target. According to the Fed’s economic projections released in March, the central bank is about halfway through its current hiking cycle and foresees two or three additional hikes in both 2019 and 2020. At the same time, the Fed assumes that QT will run in the background without disrupting markets.
The policy rate and r*: in search of neutral
The policy rate was the Fed’s tool of choice to begin the normalisation process due to the greater familiarity and extensive experience with it compared to the bank’s balance sheet.
Since 2012, the Fed has provided guidance on the evolution of the policy rate, the so-called dot plot, in its quarterly Summary of Economic Projections. Each dot on the chart represents a FOMC member’s view on where the policy rate should be over different horizons.
The latest dot plot (see chart 1) showed that the median forecast is for two more hikes this year (taking the total to three for 2018), three next year and two in 2020. The policy rate is expected to reach 3.375% in 2020 – somewhat above the current FOMC long-term rate estimate of 2.875%. This suggests that the FOMC expects its monetary policy stance to turn slightly restrictive towards the end of the forecasting horizon.
Chart 1: The Fed’s March dot plot
Source: Federal Reserve and Summary of Economic Projections as at 31 March 2018. Note: The midpoint of the target range or target level for the federal funds rate (median) is marked in orange.
The Fed’s ability to deliver on its interest rate hike plans is unclear. Indeed, since the introduction of the dot plot, the Fed has tended to revise down its expectations of the future path of the policy rate. In doing so, it has gradually reduced the envisaged pace of tightening and the expected terminal rate. This process reflected the evolution of their assessment of the neutral or natural rate – the so-called r*, the equilibrium rate when the economy is at full employment.
Unfortunately, r* is not observable and there is considerable uncertainty surrounding its estimate. Judging by the low realised GDP growth rates (probably reflecting a decline in trend growth), the real natural rate has descended in recent decades. Based on extensive research2 by Thomas Laubach and John Williams3, the current estimate of r* runs slightly above zero, compared to a pre-crisis estimate of around 200 basis points (bps).
This implies that, at the current stage of normalisation, the policy rate is probably fairly close to neutral. Using the personal consumption expenditures (PCE) price index as the deflator, the real policy rate is currently in negative territory by between 25 to 50bps – only two hikes away from a return to the current Laubach-Williams estimate of neutral (see chart 2).
Chart 2: The Fed’s real policy rate vs Laubach-Williams r* estimate
Source: Federal Reserve Bank of San Francisco, Federal Reserve, and Bureau of Economic Analysis as at May 2018. Note: The Fed’s real policy rate is deflated by PCE inflation.
As the policy rate is approaching prevailing estimates of neutral, discussion in the Minutes of the May meeting suggested that the forward guidance on rates – stating that monetary policy would remain accommodative in order to sustain a return to 2% inflation – should be removed or adjusted. That echoed recent comments by San Francisco Fed President John Williams.
That said, the neutral or natural rate evolves over time as secular forces affect the economy. In particular, the Fed assumes that r* will increase somewhat going forward, reflecting stronger trend growth as legacy issues from the financial crisis, which have been headwinds, dissipate over time. The long-term nominal rate of 2.875% in the latest Fed’s dot plot implies that the real natural rate is expected to increase to about 90bps over time. From a historical perspective, that is still a low level, reflecting a negative outlook for productivity growth and demographics4.
In general, uncertainties about the estimate of r*, its evolution, and future economic developments mean that the FOMC’s projections for the policy rate, as depicted by the dot plot, should be taken with a pinch of salt, particularly towards the end of the forecasting horizon. While the Fed will probably deliver two or three additional hikes this year, a persistently low r* and economic disruptions might lead to slower and limited interest rate hikes in coming years.
Shrinking the Fed’s large balance sheet
The assessment of the Fed’s monetary policy stance is complicated by its large balance sheet (see chart 3). According to our estimates, the outstanding amount of QE, at $4.4tn, is equivalent to a 550bp easing in policy rate terms.
Chart 3: The Fed has begun to trim its $4.4bn balance sheet
Source: Federal Reserve and Bureau of Economic Analysis as at May 2018.
The Fed has been somewhat dismissive of the impact of the QT programme, which began in October 2017. As suggested by Fed Chair Jerome Powell recently, it is assumed that the once-dreaded taper will run quietly in the background, with little impact on financial markets and the real economy, given the gradual and predictable nature of the plan.
The reinvestment reduction started at $30bn a month and will increase until it reaches $50bn a month by Q4 2018 (split 60:40 between Treasuries and mortgage backed securities). If the QT programme proceeds accordingly, the Fed’s balance sheet will shrink by about $1tn by the end of 2019 (see chart 4). According to our estimates, that is equivalent to an additional 130bps of tightening, i.e. about five additional rate hikes over the next year and a half. However, as the policy rate is nearing the neutral level and the US business cycle is probably mature, it is unclear whether the economy would be able to seamlessly absorb the projected double tightening – from policy rate hikes and balance sheet reduction – in the next couple of years.
Chart 4: The Fed’s balance sheet reduction plan
Source: Hermes Investment Management, based on the FOMC June 2017 Addendum and September 2017 Statement
Beyond that horizon, however, there is little clarity from the Fed about what balance sheet size it deems normal. Fed officials have been vague about the level they are targeting in the medium to long term.
Much of the uncertainty stems from the level of excess reserves needed to accommodate fluctuations in autonomous factors and banks’ demand for reserves. Excess reserves are moneys sitting idle on the Fed’s balance sheet that did not find a way into the real economy. A reduction of excess reserves, therefore, should not have a significant impact on the real economy in theory. However, regulation and a more prudent approach by banks might stoke higher structural demand for reserves, which in turn could impede a significant and smooth reduction of excess reserves.
In general, Fed officials concede that the central bank’s balance sheet will stabilise at higher levels than those prevailing before the financial crisis. Furthermore, they acknowledge that there is significant uncertainty about the level of excess reserves and the eventual size of the Fed’s balance sheet at the end of this cycle. For example, New York Fed President Dudley suggested that a “new normal” for the size of the Fed’s balance sheet is probably between $2.5tn and $3.5tn – Powell also alluded to this range of estimates. In September 2017, Dudley said:
“Together, as a rough starting point, we have suggested that the necessary amount of excess reserves could be in a range of $400 billion to $1 trillion. Coupled with uncertainty about the likely growth in other factors, such as currency outstanding, this implies a normalized balance sheet size of, perhaps, $2.4 trillion to $3.5 trillion in the early 2020s5."
Bumps, twists and turns on the path to normalisation
Several potential developments could disrupt the path to monetary policy normalisation, thereby accelerating or dragging out the process.
The current normalisation cycle has already hit a speedbump. In 2016, the global economy experienced a negative mini-cycle, which prompted a year-long pause in the normalisation process. Going forward, the Fed’s course towards normalisation could be affected by: the significant fiscal stimulus announced by the US administration, US yield curve developments, a possible overshoot of inflation and, most importantly, the potential escalation of protectionist threats. On balance, we believe that existing risks imply that a slower normalisation process is more likely than a faster one.
Powell has usually skirted questions about fiscal policy effects on the Fed’s normalisation process, maintaining that monetary policy remains the Fed’s sole focus. However, there are at least a couple of channels through which fiscal stimulus might affect monetary policy. First, the significant fiscal stimulus announced by the US administration – including both tax cuts and an increase in spending caps – comes at a time when the US business cycle is mature and the US economy is probably back to about potential. As the announced measures (which are likely to boost growth by about 0.5 percentage points in both 2018 and 2019) are unlikely to improve the economy’s supply-side significantly, they pose a risk of overheating. And although there might be a case for monetary policy to offset the effects of fiscal policy, the uncertainty about its actual impact means that the Fed is unlikely to take action in the short term (see chart 5).
Chart 5: The US policy mix is still accommodative, courtesy of fiscal easing
Source: Hermes Investment Manager based on data from the Federal Reserve, Congressional Budget Office, and Bureau of Economic Analysis as at May 2018.
At the same time, the announced fiscal stimulus will lead to a significant increase in the supply of Treasuries. Deficit funding needs will amount to about $2.1tn cumulatively in the next two years (see chart 6). In addition, the Fed’s balance reduction is worth another $500bn in Treasuries, implying net supply to the market will stand at about £2.6tn in 2018-2019. In GDP terms, US funding needs in the bond markets are set to be the largest among developed markets, excluding Japan. A lower appetite for Treasuries in the face of greater supply could result in tighter financial conditions – that’s something the Fed might need to take into account.
Chart 6: Treasury issuance is expected to reach record levels in the next two years
Source: US Treasury and Hermes estimates as at May 2018
Another issue that most Fed officials have tended to downplay is the impact of further rate hikes on the US yield curve. The US yield curve has flattened significantly over the last year and the spread between 10-year and two-year yields is now running at about 50bps. The Fed might find itself in a situation where additional rate hikes lead to a curve inversion – a situation where short-term rates exceed long-term yields –in the not so distant future. That’s because moves in the policy rate primarily affect the short end of the curve.
As discussed in our February commentary, ‘The US yield curve as a predictor of recessions: should we trust it this time?, a curve inversion has accurately predicted recessions in the past (see chart 7). For this reason, the Fed will think twice before actively pushing the yield curve into inversion.
Chart 7: Curve inversions have accurately signalled imminent recessions
Source: Reuters Datastream as at May 2018. Note: recessions are denoted by the shaded periods.
Another risk that could materialise is a rise in inflation above the official 2% target, given the tight US labour market. However, our view is that, as long as the inflation overshoot is temporary and contained, the Fed is likely to stick to a gradual approach, avoiding a more hawkish stance.
Indeed, there is an argument for tolerating an overshoot in inflation for a period: it would make up for the persistent undershoot in inflation in the years that followed the financial crisis. The same point is made in the minutes of the May FOMC meeting:
“It was also noted that a temporary period of inflation modestly above 2% would be consistent with the Committee’s symmetric inflation objective and could be helpful in anchoring longer-run inflation expectations at a level consistent with that objective6."
All things considered, the main risk clouding the outlook for the US and the global economy is still the high likelihood of policy accidents for global trade. In particular, an escalation of protectionist measures would act as a supply shock, resulting in a period of higher inflation and slower growth. The Fed would probably look through a temporary surge in inflation, as the impact from slower growth would eventually prevail. Protectionist tensions are unlikely to fade anytime soon. Going forward, a more confrontational relationship between the US and China will probably become a new feature of the geopolitical landscape, and the risk of an escalation will probably re-emerge on a regular basis.
Change at the top: how new personnel will impact the existing policy framework
A more fundamental change in the Fed’s approach could stem from recent changes in its personnel. Not only did Powell replace Janet Yellen as FOMC Chair in February this year, but US President Donald Trump also appointed four new Governors: Randal Quarles, Richard Clarida, Michelle Bowman and Marvin Goodfriend. Quarles, the Fed’s vice-chairman for financial supervision, is already participating in FOMC decisions, while the soon-to-be Fed vice-chair Clarida and board members Bowman and Goodfriend are still waiting for confirmation from the US Senate. In addition, there have been few changes at regional Federal Reserve Banks. Most notably, San Francisco Fed President John Williams will replace William Dudley as New York Fed President, after the June FOMC meeting7. The position is one of the most influential for shaping monetary policy as it enjoys a permanent vote on the FOMC.
Our overall assessment is that new personnel at the Fed will preserve policy continuity in the short term – even though some have voiced hawkish views since their appointment. The new FOMC members have an orthodox attitude towards monetary policy, and they are unlikely to disrupt a normalisation plan, that has been successful thus far. Only a change in circumstances would bring about an adjustment to the policy course, which would be facilitated by the pragmatic and collegial approach of Powell.
The implications of the new FOMC composition could be potentially more significant in the medium term, especially when the debate on the evolution of the practice of monetary policy has intensified. The discussion has recently focused on how to revive the existing monetary policy framework in a context of structurally low interest rates, where the zero-lower bound is more constraining. Thus far, arguments have been made in favour of adopting higher inflation targets or a price level targeting approach (see, for instance, Williams8 and Bernanke9).
The arrival of new FOMC members with different backgrounds could shift attention to different potential evolutions of monetary policy. Notably, there might be a greater emphasis on a rule-based approach à la John Taylor, which would point to a much higher policy rate (as high as 4% assuming a 5% non-accelerating rate of unemployment (NAIRU), or closer to 3% assuming a 4% NAIRU). Moreover, some new appointees have been critical of the Fed’s unconventional monetary policy tools (notably, QE), stressing the potential unintended consequences in terms of financial stability risks.
The road ahead
Today, the Fed faces what looks like an easy task – to continue along its gradual path towards normalisation. The US economy is expanding at a solid rate, the labour market continues to improve and inflation has increased towards the 2% target. The Fed’s normalisation plan predicts two to three more hikes a year until 2020, as well as the ongoing, gradual balance sheet reduction.
However, the Fed cannot be complacent. In the short to medium term, the significant fiscal stimulus announced by the US administration, US yield curve developments, a possible overshoot of inflation and the potential escalation of protectionist threats might interfere with the normalisation process. As protectionism is probably the most significant threat, the normalisation process might turn out to be slower rather than faster.
In the longer-term, the new post-crisis regime poses a new set of challenges: lower trend growth implies that the end point for the policy rate is way below its historical average. Moreover, the Fed’s balance sheet will also remain well above pre-crisis levels. In this context, the lower bound for rates is more constraining, leaving the FOMC with limited ammunition to respond to the next crisis. A new monetary policy framework may, therefore, be warranted – and perhaps then, the new FOMC members will shape the debate.
- 1 “Minutes of the Federal Open Market Committee May 1–2, 2018”, published by the Fed on 23 May 2018
- 2 "Measuring the Natural rate of Interest", by Thomas Laubach and John Williams, published by Review of Economics and Statistics in November 2003.
- 3 Thomas Laubach is an economist on the Board of Governors of the Federal Reserve System. John Williams is the president of the Federal Reserve Bank of San Francisco. He will replace Bill Dudley as New York Fed president on his retirement on 17 June.
- 4 For a recent discussion on the drivers of r*, see “The Future Fortunes of R-star: Are They Really Rising?”, speech by John Williams, published by the Federal Reserve Bank of San Francisco on 21 May 2018
- 5 “The US Economic Outlook and the Implications for Monetary Policy”, by William Dudley, published by the Federal Reserve Bank of New York on 7 September 2017
- 6 “Minutes of the Federal Open Market Committee May 1–2, 2018”, published by the Fed on 23 May 2018
- 7 The FOMC Governors, the NY Fed President and four rotating President from the Regional Banks (out of a total 12) have voting status at policy meetings.
- 8 “Preparing for the Next Storm: Reassessing Frameworks and Strategies in a Low R-star World”, by John Williams, published by the Federal Reserve Bank of San Francisco on 8 May 2017
- 9 “Monetary Policy in a New Era”, by Ben Bernanke prepared for the conference on Rethinking Macroeconomic Policy, Peterson Institute, in October 2017