An interest rate rise in the US is likely sometime this year and most private sector economists expect this to happen on September 17th. But what should the Fed should consider before making its move? It may be worth reviewing the Fed’s mandate – ensure “maximum employment, stable prices and moderate long-term interest rates” – in order to answer this question. I argue that forecasting maximum employment – or the NAIRU – could be its most important task.
Maximum employment is a somewhat vague term. It is unobservable, hard to describe and has other names: full employment, the natural rate of unemployment or the non-accelerating inflation rate of unemployment (NAIRU). My best attempt at describing it is the level of employment (which can be measured by the unemployment rate) where supply and demand for workers is balanced. If the supply of workers exceeds demand for workers, wages will be low. When demand exceeds supply, wages rise.
The problem the Fed faces is that maximum employment is very hard to estimate, not least since the skills of workers and demand from firms is constantly in a state of flux. The Fed controls the demand for workers through setting interest rates, but how much it should increase/decrease demand? There are lags between when it takes actions (like raising interest rates) and when the effects are felt on the real economy (i.e. employment and inflation), so estimating the NAIRU (and when it is likely to be reached) is key.
And the Fed’s recent record has been poor. The NAIRU it is notoriously hard to predict. It is higher in Europe than the US – perhaps because of a more generous welfare state in the former? It is lower in Japan, probably, in part, due to an older population (older people are less likely to be unemployed than younger people). Note: the Japanese unemployment rate is 3.6% yet deflation is still a threat/problem.
The Fed estimates the NAIRU to be 5.0% and 5.2% even though the unemployment rate is currently 5.3% and inflation is just 0.3%. Low inflation is partly due to oil prices over which the Fed has no control. However, core inflation (i.e. excluding food & energy) is still only 1.3%, significantly below its target of 2 percent. Note: the inflation target of 2 percent is what the Fed determines fulfills the second part of its mandate [price stability]. So if the economy is so close to full employment, then why is inflation so far away from target?
Some will say that this is because the unemployment rate is currently distorted. For example, the number of workers that are employed part-time, but want a full-time job is high by historical standards. Also, some people want a job but have become so discouraged that they haven’t been looking for work – the number of people in this situation is a lot higher than usual too. They are not counted as unemployed since they are not in the labour force. If included, the unemployment rate would be higher. Presumably they will re-enter the labour force as the labour market improves.
The Fed has alluded to this point, which the data supports. However, I reckon that the Fed’s estimate of NAIRU is still wrong. This has systematically been the case for the last few years. In June 2013 Ben Bernanke described low inflation as “transitory”. He was wrong – core inflation has remained well below target since. The Fed’s estimate of the NAIRU has been revised down and could be revised down further still: 5.2% – 6.0% in June 2013; 5.2% – 5.8% in September 2013; 5.2% – 5.6% in March 2014; 5.2% – 5.5% in June 2014; 5.0% – 5.2% in March 2015.
Before the crisis, the CBO and others estimated the NAIRU to be 5.0%. Most economists thought that following the sub-prime crisis that the NAIRU actually rose. I believe the argument goes like this: since long-term unemployment rose, workers’ skills deteriorated, thereby decreasing the supply of suitable workers for jobs. By extension, wage pressures would rise with a relatively higher unemployment rate.
Of course, this did not happen. It is true that long-term unemployment is high and the skills of these workers may have deteriorated. But, how skilled or experienced do you need to be to drive and Uber taxi, or work at McDonald’s, or work in a restaurant, or do stuff on TaskRabbit? More skilled workers tend to be able to reinvent themselves if necessary anyway. I don’t buy that argument and the wage data clearly suggests that the NAIRU has not be reached.
In fact, I believe there are a few reasons why the NAIRU is below the Fed’s current estimate, not least because of technology and demographics. In some industries technology competes with human beings, lowering demand for workers (or lowering the wages offered). For example, cashiers have been replaced by machines at supermarkets. Taxi drivers could soon be replaced by driver-less cars if Google, I mean err Alphabet, is successful. Software such as Uber has pushed down the wages of black cab drivers (although it may have pushed up the wages of Uber drivers who may have been doing something with lower pay before).
But technology is not just a one-way ticket to lower wages.Technology creates and destroys jobs. Creative destruction. In the past I think its clear that technology has created more jobs than it destroys (artificial intelligence could be a different story though I admit) – a hundred years ago, a third of the US population was employed in agriculture, now that is just 2% [people have found jobs in other industries following the technological revolution in farming techniques]. In some industries technology compliments workers and makes them more efficient, thus increasing the demand for workers and wages. For example the invention of drills, cranes and diggers complemented construction workers; output and employment in the sector rose has risen over the last 50 years due to greater worker efficiency.
It can also be argued that technology has decreased “frictional unemployment”, the name given to unemployment resulting from the time it takes job seekers to find a suitable job. Some say that online job forums such as LinkedIn now make it easier to match employers with potential workers (which reminds me I need to cancel my LinkedIn premium subscription which I have found pretty useless). I’m not entirely convinced, since too much choice can sometimes be a bad thing and result in less efficient matches. Think Tinder.
The US population is ageing, which naturally leads to a lower natural rate of unemployment. University enrollment is rising, also leading to a lower NAIRU. Rising inequality could also be lowering the bargaining power of workers in wage negotiations. The San Francisco Fed has said that there was pent up slack following the crisis – firms did not cut wages as much as they would have given the drop in demand during the crisis years (sticky wages both up and down).
And I think that below target inflation is likely to continue. And not just because the natural rate of unemployment is lower than thought. Commodity prices could well stay low given the economic slowdown in China. This doesn’t only affect the headline inflation, it gets passed through to core inflation too! The US Dollar is strong so imports are cheap. China has devalued its currency, which will further contribute to low inflation in the US. Technology spending is also at record highs and is likely to lower all kinds of costs: transport, communication, medical. Low inflation and low NAIRU would appear to suggest that the Fed should refrain from raising rates.
I would say on balance yes, but there are a few additional factors to consider that would support raising them now. Firstly, interest rates are currently very low by historical standards. So, even if the Fed does raise rates now, policy will still be accomodative. Second, there are lags between when policy decisions are made and the effects on the economy are felt. The Fed needs to predict how close the economy is to the NAIRU is six months or one year’s time. Third, if it waits till inflation starts accelerating, then it will need to tighten aggressively – this could tip the country back into recession. In other words, it could start now in order to ensure a smoother upward path. Another argument is financial stability: low rates fuel bubbles in asset markets. This is a weaker argument. Besides, the Fed has new powers to regulate banks and ensure financial stability. It has also said that interest rate policy would be a blunt tool for ensuring financial stability. One might also add – if your name is John Maynard Keynes – that raising rates to stabilise the financial system is tantamount to curing the disease by killing the patient.
Anyhow, what the Fed should do and what it will do are two different things. Janet Yellen and other FOMC members have said that they want to be “reasonably confident” that inflation will move back towards target before raising rates. But I don’t really know how they can be – wages haven’t risen and inflation is not showing any signs of moving back towards target. 80% of economist are predicting a rate hike in September. I’ll be different and say October.
Note: these views are mine, and mine alone. They do not reflect those of my employer (past, present or future) or anybody that has or will pay me money. They may be inaccurate and you should not rely on them, at all. They are also subject to change, at the drop of a hat.