Powell Rules

There’s a new kid in town. Jay Powell’s first major speech as Chair of the Federal Reserve, the Monetary Policy Report to Congress, had an immediate impact on markets, interpretated as indicating a more hawkish stance than his predecessor. Time will tell but one interesting feature of his speech was his emphasis on monetary rules in setting policy, which  he finds ‘helpful’, with an analysis of five such rules detailed in the Report.

It seems clear that the Fed do not slavishly follow any rigid precription in setting interest rates, although the rate paths implied by the various rules  are apparently set out ahead of FOMC meetings in order to act as a guide, and Powell’s speech is likely to stimulate further market interest in this area.

The best known rule is named after John Taylor , and posits that the Fed funds rate should move by precribed amounts from its long run equilibrium level  ( which on current FOMC forecasts is 2.75%) if inflation differs from the 2% target or if the real  economy  has moved away from its full employment level, which the Fed currently believes is consistent with an unemployment rate of 4.6%.

What does the Taylor rule imply now? The Fed expect inflation in 2018 to pick up to 1.9% but that the unemployment rate by the final quarter of the year will have fallen to 3.9% and so below the long-run equilbrium level , indicating that tighter policy is required, with an implied  Fed funds rate of 3.3%, which is around 1% higher than the median FOMC expectation as set out in the ‘Dot Plot’.  In other words rates would rise more rapidly than curently envisaged by the market , although over the following few years the ‘Dot Plot’ converges to the Taylor rule, albeit with the latter implying a modest easing of policy while the former points to a steady tightening. The Taylor rule also implied the need for negative rates following the financial crash and an “Adjusted Taylor rule’ take account of this insufficient monetary accommodation in the past by advocating a gradual return to the rate implied by an unadjusted Taylor rule, although the former has now largely converged to the latter.

What of the other rules discussed? A ‘balanced Approach’ rule gives a greater weight to deviations from full employment and that also indicates that policy is too accommodative, and indeed should be much tighter by end-year, with a Fed funds rate of 4%, before falling back to 2.75% in the long run.  Not all rules imply the Fed is behind the curve, however, with the other two rules discussed arguing against aggressive tightening. One, the ‘First Difference rule’ takes account of the current level of the Fed funds rate and the pace at which unemployment is changing, and implies a policy rate of 1.75% by end-year, which is below the ‘Dot Plot’ figure . Similarly, another variant, ‘the ‘Price Level rule’, implies that policy is also too tight now and as projected because the rule adjusts for the fact that inflation has been below the 2% target for some time, so the price level is therefore lower than would be the case had inflation been at 2% every year.

All such rules are based on simplified models of complex  and changing dynamics in the US economy, but ,as Powell noted, can be useful for policy makers . Three of the five imply that rates are too low given the Fed’s expectations for inflation and unemployment which may prove more significant with Powell at the helm.


Are Inflation targets too high?

Inflation targeting by policy makers emerged in the early 1990’s and is now part of the standard toolkit for most central banks, with operational independence from government also the norm. The idea is straightforward; if the central bank commits to hitting a specific inflation rate, that rate will impact expectations and eventually will help to anchor price changes. Too high a figure means that purchasing power is eroded at an unacceptably fast pace, while too low risks periods of deflation.

The latter view persuaded central banks to eschew a zero inflation target, and the figure of 2% is very common, although of course it means that prices rise by 22% over a decade and by 50% in a generation.  Hardly stable prices then, although one should remember that such targets  were often set some time ago and at a period when inflation was generally above that figure.In contrast, many central banks have been wrestling with the opposite problem for some time i.e. inflation is persistently below target.

In the US, for example, core inflation ( the CPI excluding food and energy) has been below 2% for almost a decade, while in the euro area the last time core inflation exceeded 2% was back in 2003. This points to structural factors at play, rather than purely cyclical drivers.

Standard inflation models , however, generally posit a cyclical  link between economic activity and inflation, with periods of stronger growth resulting in an acceleration in inflation. The link is often based on the Phillips curve, the idea that falling unemployment will boost wage growth and hence lead to price rises. Consequently, most central banks expected inflation to pick up given falling unemployment, particularly as rates for the latter are now very low by historical standards in some countries, including the US and the UK. Yet wages have not picked up as expected ( the Phillips curve has flattened), reflected a range of factors, many of them structural, including globalisation and free trade, a shift in employment composition to lower productivity jobs, the rise in self employment,  fear of job losses and the decline in trade unions .

There may be other factors directly impacting inflation, such as the greater ease of price discovery in a digital age and the growth of disruptive technology, ( examples might be Amazon and Uber) which are replacing traditional models of distribution. Technology change is general is also shifting the aggregate supply curve rightward, so putting downward pressure on prices.

Yet the Fed and the ECB are still wedded to a 2% inflation target, despite missing it to the downside for a long period. Fed chair Yellen did acknowledge recently  that inflation was not behaving as expected ( calling the recent inflation performance ‘a mystery’) but the FOMC and the ECB are both of the view that cyclical factors will eventually win out, pulling inflation up to target.They may be ultimately proven right, although the recovery is now pretty long in the tooth, particularly in the US, and the Fed has revised down its view of long run potential growth ( now sub 2%) although not of long run inflation, still at 2%.

Does it matter if inflation was to stay in a 1%-1.5% range. It’s obviously better than 2%  for many people in an era of weak wage growth but it does raise a policy issue- if ‘equilibrium inflation’ is now below 2% due to structural changes, then policy will be too loose if it is set to hit 2%. The liquidity currently flooding the world has indeed driven up prices, but equities and property rather than the price of goods and services.


Fed Hard to Fathom

In December last year the US Federal Reserve tightened monetary policy, albeit by only a quarter point, citing the ‘considerable improvement in labour market conditions’ and an expectation that inflation would gradually recover to the desired 2% level. The accompanying statement emphasised that further rate increases would be gradual, although the projections released at the time from the 17 FOMC participants (the ‘dot plot’) indicated a median expectation  that 2016 would see four further quarter point rises.

The market was not convinced but the latest ‘dot plot’ , released in mid-March, was still a surprise to many, as the median expectation now showed only two rate increase by year-end. True, growth was now projected to be marginally weaker in 2016 and 2017 but the unemployment rate was also forecast to be  lower, falling to 4.5%, against  4.8% in the long run. Inflation ( the Fed’s measure is the personal consumption deflator)  was expected to end the year at 1.2%, down from the previous 1.6% forecast, but the core rate forecast  (defined as ex food and energy) was unchanged at 1.6%.

Monetary policy is deemed to have ‘long and variable lags’ so one might think that unemployment around the Fed’s desired level alongside an expectation that inflation will pick up would be consistent with steady rate increases. let alone a change to a more dovish rate outlook. Core inflation has actually  picked up of late , rising at an annual 2% rate over the past three months, and the annual rate is 1.7%, which is above the Fed’s expectation for the year-end. Yet Janet Yellen, in a recent speech, emphasised that low inflation expectations were now a concern. Survey measures had shown a fall while market based measures had also declined ; the 10-year break-even inflation rate fell to 1.2% in February. She highlighted the risk that lower expectations would feed into wage and price setting, so increasing the probability of inflation remaining below target for longer. The Fed Chair also flagged the possibility that  the unemployment rate could fall much further without triggering wage inflation and  also pointed to international risks to the US economy.

So it is now very difficult to see what would trigger a rate rise in the coming months. A further improvement in the labour market does not seem a sufficient condition and higher spot inflation may not be sufficient either, given the emphasis put on inflation expectations. In fact the latter is heavily influenced by spot inflation anyway, particularly high frequency purchases like fuel, and  market expectations have actually risen again of late, perhaps driven by the rebound in oil prices in March.  Crude prices have  fallen back again in the past week, however, and this may dampen expectations again.

The Fed funds future for December is trading at 0.5%, so the market is no longer convinced we will see any rate increase this year. Central banks generally tread a line between Discretion on policy or following a more Rules based approach and the former is driving the Fed at the moment, which makes for flexibility but makes it very hard to read their next move.