Fonds im Fokus


15.12.2017 | 12:37

NN IP: Fed policy will become less straightforward

As expected, the Fed hiked rates for the third time this year. Going forward, Fed policy is set to become less straightforward as we move further into 2018, also because the central bank will have to decide how to respond to the tax cut and because policy is getting closer to neutral.

The Fed will stick to the “golden mean” policy route…

The Fed announced its widely expected third rate hike of the year yesterday. Going forward, the policy rate path is far from certain. The Fed is trying to navigate three cross currents and may sooner or later have to make a basic choice which one of these it deems to be the dominant one.

The first cross current is an underlying growth momentum in the 2-2.5% range (and a bit higher than that in Q2/Q3 due to one-offs such as inventories) coupled with a declining trend in the unemployment rate which has now fallen towards 4%. The second cross current can be summed up by the notion that 8 years into the expansion core PCE inflation (1.4% at the latest reading) is still running well below the 2% target, even though its short-term momentum has improved somewhat. Meanwhile, in the labour market wage growth also remains surprisingly weak, even if one adjusts for low inflation and low productivity growth. The third cross current is a substantial easing of financial conditions, despite four rate hikes since December 2016. At the very least, this easing of financial conditions is providing more impetus to the first cross current. According to some estimates, this is a boost to 2018 growth of 0.5-1pp relative to the baseline where financial conditions had remained at their December 2016 levels. This seems a bigger boost than we can expect from the tax cut! At its worst, the easing of financial conditions could eventually lead to bubbly conditions in some market segments, even though the Fed and ourselves are not yet worried about that. An important reason for this is the fact that household and bank balance sheets look a lot better than they did in 2008. At any rate, the easing of financial conditions bears the footprint of monetary easing (dollar deprecation, stable to lower US Treasury yields and rising equity markets) which suggests that r-star could well have risen on the back of a stronger feedback loop between growth momentum and financial conditions, which is ultimately underpinned by an improvement in animal spirits.

Amidst these cross currents gradualism may then well seem like the “golden mean” policy route. In fact, the Fed’s statement of longer-run goals and policy strategy holds that when the Fed faces a conflict between the employment and inflation sides of its mandate, it will take a “balanced approach”. The latter aims to minimize deviations of both objectives from their respective targets. Actually, this still requires the Fed to tell the markets what the relative weight on each of these objectives is. The term “balanced” suggests that each objective gets equal weight. However, it is far from sure that this will always be the case. Suppose the Fed is dealing with a 10% unemployment rate and a core PCE inflation rate of 2.5% which for some reason (e.g. a persistent dollar depreciation trend) is not expected to fall much over the next year. However, inflation expectations remain firmly anchored to the target. In such a situation the deviation of the unemployment target is much larger than that of the inflation target, so the Fed is likely to put a bigger weight on the former. Now imagine the same situation but inflation expectations are in grave danger of ratcheting upwards. In that case the Fed is likely to give a very big relative weight to its price stability objective.

The upshot is that the golden mean policy route is not straightforward even in a simple theoretical world where the central bank only tries to reduce the volatility of unemployment and inflation around their respective means (NAIRU and the inflation target). In the real world, additional complications can arise and the current situation is a perfect example of this. On the one hand, you could argue that two cross currents (labour market tightening and financial stability risks) entail a risk of “undesired consequences” from being too dovish ex-post. Meanwhile, only one cross current (below-target inflation) could get worse if policy turns out to be too tight in the end. This seems to be more or less the line of reasoning of the hawks on the FOMC where they add a good deal of faith in the Phillips curve in the mix. On the other hand, one can argue that the anchoring of inflation expectations is the ultimate goal of monetary policy because of which the price stability objective should get a much larger weight than the full employment or financial stability objectives. After all, inflation will tend to be mean-reverting if and only if inflation expectations remain stable at 2%. This allows monetary policy much more room to respond to shocks compared to a situation where expectations become unmoored.

This holds especially because unmoored expectations have a tendency to drift away further in the same direction. The reason is that agents then start to use inflation rates from the recent past as benchmark for expectations. This tendency may be even more pronounced for a downward drift in inflation expectations than for an upward drift. The reason is that the distribution of macro outcomes is skewed to the downside and has “fat left tails”. Also, the lower inflation expectations fall the less room there will be in future to cut the real policy rate. This is a feature which may contribute to a further decline in inflation expectations. Hence, in this view of the world (advocated by Brainard, Kashkari and Evans) the Fed should not be raising rates at all until it sees the “whites of the eyes” of inflation which in practical terms means until it has seen a clear overshoot of the target. In the eyes of these dovish folks the very fact that the Fed is hiking in an environment where there is a substantial risk that inflation expectations have slipped below the target could potentially act as a negative signal which reinforces the trend in inflation expectations.

…but this could still lead to suboptimal outcomes

Regular readers will know that we have some sympathy for the latter view. Still, our job is to predict what the Fed will do and not what we think the Fed should do. The core of the FOMC is clearly firmly wedded to the balanced approach which pays attention to all three cross currents. Having said this, some little cracks have started to appear in this consensus. We do not think they will become big enough to cause the Fed to deviate from the current game plan in the future, but it may very well lead to a strategic reassessment of its goals which we will discuss at a later stage. Recent remarks by Yellen are pretty revealing in this respect, even though they may need to be taken with a pinch of salt. With only a few months to go as Fed Chair, she may be more inclined to express her own opinion than to “toe the Fed party line”. Still, her doubts may well be representative for more FOMC participants. Last month, Yellen confirmed that the 2017 inflation trajectory is a mystery to her. Undershooting in 2014-16 could be well explained by lingering high unemployment, falling oil prices and a sharp dollar appreciation. By contrast, all these fundamentals pointed to a pick-up in inflation momentum in 2017 which did not occur. Yellen still expects that transitory factors are an important part of this story and that these factors will wane in 2018 but admitted that “there may be something endemic here that we need to pay attention to.” The pink elephant in this sentence is obviously the possibility of slipped inflation expectations.

What’s more, Yellen even hinted at the possibility that the two macro stability goals, i.e. on target inflation and maximum sustainable employment, may not even be in conflict with each other. At the ECB policy conference in November she said, “Is it inherently bad that unemployment is below the natural rate? You know, not sure”. This statement brings back memories from last year when she encouraged economic researchers to delve into the issue of running “a hot economy” with the aim of pushing the supply side onto a permanently higher path. At the time, the market took this as a dovish statement and Yellen had to publicly backtrack subsequently by stating that the Fed did not have the intention of deliberately overheating the economy. Presumably, running the economy hot could well be an idea which appeals to Yellen but does not fit with the consensus opinion on the FOMC. Hence, once again we think Yellen betrayed a personal preference here but the situation is different than last year. When she talked about running the economy hot the unemployment rate was still around 5%, which was still somewhat above the median NAIRU estimate of 4.6%. In the meantime, the Fed has more or less silently moved towards running the economy hot as the unemployment rate is now 4.1%. We have to make one qualification here. The Fed is now running the economy hot when we take its own ex-ante NAIRU estimate as a benchmark. There is still a substantial possibility that the Fed is not running the economy hot when taking the actual unobservable NAIRU as a benchmark. The behaviour of (wage) inflation acts as the smoking gun which points to this possibility.

After all, the behaviour of inflation is the ultimate arbiter of whether the economy is hot or not. In fact, the very reason the Fed does not want to run the economy too hot for a long time is precisely because this will trigger a sharp acceleration in inflation, which will force it to hit the monetary brakes so hard that a recession becomes inevitable. If it is indeed true that the actual NAIRU is below 4%, there is no conflict (yet) between the Fed’s twin objectives. Some pundits may think that it is ludicrous to believe that the NAIRU could be anywhere below 4% because two or three decades ago estimates were in the 5.5-6% range. This only goes to show how fluid this concept actually is over time. The NAIRU clearly is not some immutable constant of economic nature. In fact, it is very much influenced by labour market institutions in the broadest sense of the word, i.e. the degree of unionization, ease of hiring and firing but also the behavioural aspects of wage bargaining. In a world where tastes and technologies are changing very fast and an increasingly large share of the work force is on flexible contracts, workers may be more reluctant to ask for wage increases even at very low unemployment rates. At the same time, employers may be more reluctant to grant them because they have become accustomed to a long period of low productivity growth.

A similar or even stronger type of argument can be made about financial stability. On the one hand, keeping the policy rate persistently low from an historical perspective may trigger a search for yield, which in some asset classes becomes so strong that it starts to live a life of its own. In extremis this can morph into a bubble which if it bursts can trigger a nasty recession, especially if the private sector levered up big time in the process. On the other hand, one has to bear in mind that it is far from certain that there is massive and widespread overvaluation in financial markets because a low level of r-star should reduce risk-adjusted expected returns across the entire asset spectrum (i.e. a low r-star can explain a high P/E ratio in the stock market). Also, the Fed cannot target both macro and financial stability with just one blunt instrument, which is why macro-prudential policies are the first port of call to safeguard the latter. Last but not least, a persistent drift down in inflation expectations will ultimately also be a threat to financial stability as it can cause a widespread and massive increase in debt burdens over time while pushing risk premiums in financial space persistently higher.

As a result of all this, it may be more appropriate to think of Fed policy as balancing the risks attached to the three cross currents rather than trading off an undershoot of the inflation target against an undershoot of its unemployment target. The risks involved here are as the following. On the one hand, there is the possibility that persistent lowflation will cause further slippage in inflation expectations. On the other hand, a substantial undershoot of the actual NAIRU could trigger a sharp and sudden inflation acceleration. In addition, the Fed could also trigger financial bubbles in the process. In this balancing act the Fed has already allowed the economy to overheat relative to its own metric of full employment, which is exactly why the FOMC seeks to decelerate the pace at which the economy overshoots further. This is very clear from its unemployment rate forecasts which incorporate a sharp deceleration in the pace of decline.

In the near term, this does indeed seem like the prudent thing to do but it could very well be that this “compromise” will eventually lead to a suboptimal outcome anyway. If it does, it could cause more volatility in the economy and markets than we have seen this year (even though that admittedly sets the bar pretty low). If the Fed were to decide to stop hiking until inflation is sustainably above the 2% target (as Brainard advocates) this may increase the risk of a sudden acceleration in inflation which causes an overshoot but the potential costs in term of unmooring inflation expectations to the upside seem limited. In other words, it could be worth taking this risk even though we do acknowledge that financial markets may display a knee-jerk reaction involving strongly rising Treasury yields in this scenario. Also the potential financial stability consequences can be mitigated by macro-prudential measures. By contrast, if the Fed proceeds on the current gradual hiking path it will keep the probability of future excess inflation or bubbles contained, but this will come at the cost of potential further slippage in inflation expectations. Either way, Fed policy is set to become less straightforward as we go further into 2018, also because the FOMC will have to decide how to respond to the tax cut and because policy is getting closer to neutral.

Emerging markets: again good data from China

Economic data in China continue to reflect a stable growth environment. While fixed investment growth slows moderately, export and consumption growth remains strong. Last Friday, the November trade data was published. Both import and export growth beat expectations with a sizeable margin: import growth reached 17.7% year-on-year, export growth 12.3%. The strong export growth is in line with the export growth in the other Asian countries that already reported their numbers for November. It is also consistent with global leading indicators and the November PMIs. For EM as a whole, the November export growth was the 11th consecutive month with double-digit growth.

Chinese import growth was perhaps a bit inflated by some seasonal factors that pushed the demand for commodities up for the month. But also if we adjust for this, import demand remains strong, reflecting still decent domestic demand growth. Household consumption plays a big role. Overall credit growth remains supportive as well as it is coming down only slowly. The November aggregate financing growth was considerably higher than anticipated. We do not have all credit data for the month yet to calculate the broad credit growth for November, but we estimate that growth will have been around 12.5%.

With global trade still going strong and credit growth at these levels, Chinese economic growth should stay close to current levels in the coming quarters.