Labour Markets and Monetary Policy

======================================================================

Question 1: Do you agree that a strong labour market is a good indicator of building inflationary pressure?

======================================================================

======================================================================

Question 2: Do you agree that, in a period of great uncertainty and after a prolonged period of weak real wage growth, monetary policy makers can afford to wait for greater certainty about real wage developments and building inflationary pressure before raising interest rates?

======================================================================

Labour markets, uncertainty and monetary policy: CFM Survey 

Summary 

Ahead of the May 2018 interest rate decision of the Bank of England’s Monetary Policy Committee, more than two thirds of leading economists surveyed by the Centre for Macroeconomics agree that in a period of great uncertainty and after a prolonged period of weak real wage growth, monetary policy-makers can afford to wait for greater certainty about real wage developments and building inflationary pressures before raising interest rates. The experts are more divided on whether a strong labour market is a good indicator of building inflationary pressures: just over half agree while just under a third disagree. 

Background 

The May 2018 survey by the Centre for Macroeconomics (CFM) and the Centre for Economic Policy Research (CEPR) asked the panel of top UK and continental European economists two questions about the labour market and monetary policy that are relevant to current policy discussions. Respondents were first asked whether they believe that in today’s economy, a strong labour market is less of a signal of growing inflation risks, as would be consistent with the recently expressed view that the Phillips curve has weakened permanently. 

The second question asked whether uncertainty might allow central banks greater space to wait before raising interest rates further. Directly related to this second question, after the survey was launched, Mark Carney (Governor of the Bank of England) cast doubt on the need to raise interest rates at the next meeting of the UK’s Monetary Policy Committee (MPC). He appeared to suggest that uncertainty might support more gradual increases in interest rates than were previously expected. 

Labour markets and inflation 

In the UK, as in other advanced economies, the recovery in economic activity since the financial crisis has been characterised by a marked decline in unemployment, but both nominal and real wage growth have been surprisingly weak. Two reasons for this weakness, highlighted in the May 2017 CFM-CEPR survey, are the relatively weak labour protection afforded to UK workers and the weakness of productivity growth since the crisis. 

Some economists have suggested that the Phillips curve – the negative relationship between inflation and unemployment – has weakened. This view would suggest that strong labour market data need not be taken as an indicator of building inflationary pressures. 

There are many suggested reasons for a permanent shift in the relationship between prices and labour market activity:[1]

  • Labour market changes, such as de-unionisation, may have reduced the capacity of many workers to extract substantial wage increases even if the labour market is displaying signs of an excess of demand over supply. 
  • An increasingly global marketplace and openness to foreign competition may have weakened the ability of firms to raise prices and, hence, reduced their capacity to increase wages when faced with tighter domestic conditions. 
  • Migration may also have played a role in dampening wage growth. 
  • Central banks may now have become so effective at anchoring inflation expectations that they have directly weakened the relationship between the economic cycle and inflation. 
  • In addition to anchoring inflation expectations, the Phillips curve may appear to be flat in economies in which the central bank keeps inflation in a very narrow range (though in the UK it has moved between 0% and 5% in the last ten years). 

Others argue that the relationship remains relevant. Monetary policy-makers continue to expect that a tight and tightening labour market will lead to increasing pay growth above the recent subdued trends. Higher pay is then expected to knock on to higher inflation.[2]

As Mario Draghi (President of the European Central Bank) has said, ‘As the labour market tightens and uncertainty falls, the relationship between slack and wage growth should begin reasserting itself. But we have to remain patient.’ 

Writing in the Financial Times, Gavyn Davies [3] has argued that the relationship is merely hiding due to both international forces dampening inflation and the fact that headline figures may be the wrong indicators to use to assess the relationship. 

Related to this debate, the first question in the latest CFM-CEPR survey asked panellists: 

Question 1: Do you agree that a strong labour market is a good indicator of building inflationary pressure?

Fifty-five panel members answered this question. On balance, the experts agree with the statement: 55% either agree or strongly agree; 31% either disagree or strongly disagree; and 15% neither agree nor disagree. 

Some of the respondents who agree with the statement stress a continued relationship between labour market strength and inflationary pressure: as Francesco Giavazzi (IGIER, Università Bocconi) puts it, ‘The Phillips curve is still alive’. 

Others argue that even though it has weakened, it remains a strong and important link. Fabien Postel-Vinay (University College London) writes that ‘The fact that Phillips curves seem to be getting flatter doesn’t necessarily imply that the inflation/labour market link has been severed.’ He points out that in a world of imperfect worker-job matching, there is slack when average match quality is low. Workers in a poor match are unlikely to be in a position to demand higher wages even if the unemployment rate is low. 

Many respondents felt that the link remains between the labour market and inflation, but that it cannot necessarily be measured by the unemployment rate as in the Phillips curve. Wouter Den Haan (London School of Economics, LSE) captures this view, saying ‘A low unemployment rate may not be sufficient to talk about a “strong” labour market.’ 

Potential alternative labour market metrics include measures of ‘participation rates, non-standard jobs, part-time jobs’, according to Philippe Martin (Sciences Po); and ‘the number of workers that are involuntarily part-time employed or working shorter hours than they would like’, according to Stefan Gerlach (EFG Bank). 

Christopher Martin (University of Bath) worries that even some alternative labour market indicators may be less useful now. He cites US evidence that the vacancy yield (the number of hires per vacancy) has fallen and rendered vacancies per unemployed worker a less reliable indicator of inflationary pressures. He hypothesises that a similar change may also have occurred in the UK. 

Other respondents also emphasise that low unemployment is not a good indicator but actually disagree with the statement. They include Etienne Wasmer (Sciences Po), Panicos Demetriades (University of Leicester), Pietro Reichlin (Università LUISS G. Carli) and Sir Christopher Pissarides (LSE), who writes ‘low unemployment… is not a good indicator of inflationary pressures because in the old days the inflationary pressures originated in manufacturing and unionised workers, which have become too small a group to matter.’ 

Roger Farmer (University of Warwick) and Ramon Marimon (European University Institute) both disagree with the idea of the Phillips curve. Farmer argues that even the expectations-augmented Phillips curve is ‘an irrefutable theory that contains an unmeasurable concept’, and that central banks and most practicing macroeconomists are currently working with a flawed theory. Marimon thinks that ‘even with a better measure, the causality effect on “inflationary pressure” is weak theoretically and empirically.’ 

By contrast, Ricardo Reis (LSE) considers the link between the labour market and inflationary pressure, interpreted as capturing ‘a structural relation that encapsulates the strength of nominal rigidities, monetary non-neutrality, or deviations from the classical dichotomy’, as ‘one of the most powerful, effective, and useful pieces of applied economics’. 

Monetary policy and the labour market 

Notwithstanding this debate, the March 2018 report of the Office for National Statistics on UK labour market statistics (released 21 March) was received positively by financial markets. In the three months to January, employment in the UK stood at 32.25 million, which is 168,000 more than in the previous three months and 402,000 above the same period in 2017, a larger increase than markets had expected. 

The unemployment rate also declined from 4.4% to 4.3% in January, which reversed an unexpected rise from 4.3% to 4.4% in December. Since August 2016, the unemployment rate has been below the pre-crisis value of 5.2% in January 2007.

Most interest focused on the fact that we are now seeing the first signs of an impending pick-up in real wages. First, nominal wage growth rose to 2.8% in the three months to the end of January, which is up from 2.5% in the previous three months. Moreover, the Bank of England’s February Inflation Report cited evidence from its agent surveys that are consistent with increasing pay growth. 

Second, inflation, which has been outstripping nominal wage growth for most of the last eight years, has actually shown signs of weakening. The February Consumer Prices Index (CPI) indicated that inflation has fallen to 2.7% on a year earlier. This compares with 3.0% in January, a fall that is a little more than was expected in the Inflation Report. Comparing the January nominal wage growth with the equivalent CPI inflation data suggests that real wages may have actually grown slightly. 

These labour market data moved markets such that they have now priced in an interest rate hike with more than 90% likelihood at the next MPC meeting in May. The view is that the Bank of England should further increase interest rates before wage pressures build further. 

But there is also cause for some caution. In the February Inflation Report, the Bank of England highlighted that ‘Three-month regular pay growth relative to the previous three months has remained around 3% on an annualised basis’. This measure has actually fallen from 3% to 2.5% in the most recent release. 

In addition, numerous other surveys paint a picture of stable rather than accelerating wage growth. For example, measures of expected pay for the year ahead from Confederation of British Industry surveys averaged 2.5% in the second half of 2017, which is the same as the survey for the first half of 2017. And in the second half of 2017, the British Chambers of Commerce survey of the percentage balance of companies currently facing pressures to raise prices due to pay settlements was below where it was in the first half of 2017.[4] 

Finally, the most recent decline in the unemployment rate occurred despite an increase in the number of people unemployed. There was, at the same time, a more-than-offsetting increase in the labour force. This indicates that more people are actively seeking work now than recently, but it also suggests that there may be further room for increases in the labour force. 

An alternative view is that the MPC can wait. There remains heightened uncertainty about the outcome of the negotiations surrounding the UK’s withdrawal from the European Union. Given the fragile nature of the recovery since the financial crisis, and especially the incredibly weak growth in real wages over the past eight years, the Bank of England can allow wage pressures to build further and to become more certain of the strength of the recovery and general conditions in the UK economy. 

In November 2017, two Deputy Governors on the MPC voted against the increase in the interest rate. The MPC minutes state that they saw ‘insufficient evidence so far that domestic costs, in particular wage growth, would pick up in line with the Inflation Report’s central projection’ and that ‘recent experience suggested that wage growth could continue to be less responsive to falling unemployment than past experience would suggest.’[5] 

Motivated by this recent UK data, the second question of the CFM-CEPR survey asked about the correct response of monetary policy-makers to signs of building wage pressures in the context of a weak recovery of wages since the financial crisis and continued elevated uncertainty: 

Question 2: Do you agree that in a period of great uncertainty and after a prolonged period of weak real wage growth, monetary policy-makers can afford to wait for greater certainty about real wage developments and building inflationary pressure before raising interest rates?

Sixty panel members answered this question. The experts overall agree with the statement: 71% either agree or strongly agree: 27% either disagree or strongly disagree; and only one respondent neither agrees nor disagrees. 

David Cobham (Heriot-Watt University) would wait as he questions the underlying strength of real wage growth. Panicos Demetriades (University of Leicester) stresses the Brexit uncertainty and the fact that a good Brexit deal would actually lead to an appreciation of sterling, which will help to reduce inflationary pressures. 

Tony Yates (CFM associate), Martin Ellison and Simon Wren-Lewis (both University of Oxford) all agree with the statement and argue that the downside of not raising now is small as central banks can move more aggressively once inflation starts to pick up in a more certain and sustained way. 

Gianluca Benigno (LSE) thinks that there is ‘room to wait to be certain that the real wage developments are indeed consistent’. His own recent work suggests that the MPC should wait for ‘a sustained pick-up in business investment that would result in higher productivity growth’. 

Costas Milas (University of Liverpool) notes that recent work suggests that ‘In the presence of elevated economic uncertainty, monetary policy tightening becomes less effective’, but he feels that waiting ‘until we have a clearer picture about the economy’ is justified. 

While agreeing with the statement, Jordi Galí (CREI, Universitat Pompeu Fabra) adds a cautionary note: ‘it would be a mistake to wait too long and be forced to increase rates too rapidly.’ Ray Barrell (Brunel University London) makes the same point, but it is enough for 

him to dismiss the need to wait: ‘today looks less uncertain than, say 2008-09, and no worse than average’ and ‘Uncertainty is not a reason for inaction.’ 

Joseph Pearlman (City University London) agrees with the idea of waiting but actually sees the pressure for a rate increase coming from a Schumpeterian motive. He argues that ‘it is highly likely that we need poorly performing firms to drop out of the market, and they will only do so once their rates of return are below the interest rate.’ 

Concerns about asset price inflation are enough to convince Kate Barker (British Coal Staff Superannuation Scheme) and Etienne Wasmer that rates should start to rise, gradually but immediately. As Barker says, ‘we learned about the dangers from credit imbalances in the 2000s’. Fabrizio Coricelli (Paris School of Economics) also worries about the effects of expansionary policy on asset prices, as well as the possibility that credibility could be damaged by waiting. 

Ethan Ilzetzki (LSE) feels that inflationary pressure is enough to warrant a gradual approach to interest rate increases. Jagjit Chadha (National Institute of Economic and Social Research) stresses that ‘Rates could be raised a number of times in small increments and still be providing a monetary stimulus.’ Patrick Minford (Cardiff Business School) argues that monetary policy needs to come out of crisis-mode monetary policy and so sees ‘raising rates and reducing QE [quantitative easing]’ as necessary.

[1] A selection of articles on this topic are: Summary by Bruegel (http://bruegel.org/2017/11/has-the-phillips-curve-disappeared/); Article by Larry Summers (https://www.ft.com/content/180127da-8e59-11e7-9580-c651950d3672); The Economist (https://www.economist.com/blogs/graphicdetail/2017/11/daily-chart)

[2] A UK example is Michael Saunders’ (Bank of England MPC) speech on “The Outlook for Jobs and Pay” (https://www.bankofengland.co.uk/speech/2018/michael-saunders-launch-of-financial-intermediary-and-broker-association).

[3] https://www.ft.com/content/e48ebf08-fcfe-3b82-8f91-ade8a3eada65

[4] Both as reported in the February 2018 Inflation Report in Table 4.A.

[5] Monetary Policy Summary and minutes of the Monetary Policy Committee meeting ending on 1 November 2017: https://www.bankofengland.co.uk/-/media/boe/files/monetary-policy-summary-and-minutes/2017/november-2017.pdf

Contact us for more information

How the experts responded

Labour markets and inflation

Participant Answer Confidence level Comment
Gianluca Benigno London School of Economics Disagree Confident
The recent experience in the US and the UK seems to contradict this statement. Indeed both countries have experienced a substantial decline in unemployment (if that is the metric for a strong labour market) from its peak without a significant increase in wage inflation and as such inflationary pressure have been the result of exchange rate fluctuations and commodity price fluctuations (especially for the UK economy which is a small open economy). In my opinion the main reason, from a cyclical point of view, is related to low productivity growth caused by a decline in business investment in the aftermath of the global financial crisis. There are then other forces secular in nature that are contributing in weakening the link between falling unemployment rates ad increase in inflation: mainly the decline in labour share of income and still deflationary forces coming from globalization.
Nezih Guner CEMFI Disagree Confident
Ramon Marimon European University institute and UPF-BarcelonaGSE Disagree Confident
If by "a strong labour market" means "low unemployment" I disagree in two respects. First, "low unemployment" is not a good indicator of a labour market being strong, high employment rate is a better one and even better if it takes into account of which type of employment (e.g. stable, productive, etc.). Second, even with a better measure, the causality effect on 'inflationary pressure' is weak theoretically and empirically. In any case, and in almost any measure of 'strength", in many European countries we are far from having "strong labour markets", unless 'strong' means burdensome...
Richard Dennis University of Glasgow Agree Confident
Fabien Postel-Vinay University College London Agree Confident
The fact that Phillips curves seem to be getting flatter doesn't necessarily imply that the inflation/labor market link has been severed. Traditional measures of aggregate slack focus on the unemployment rate. Yet in a world with heterogeneous jobs and workers, leading to heterogeneous "match quality" and to mismatch, slack exists also in employment when average match quality is low. Indeed, recent work (Faberman and Justiniano, 2015; Moscarini and Postel-Vinay, 2016, 2017) wage inflation comoves with the pace of Employer-to-Employer transitions, not with Unemployment-to-Employment transitions.
John Driffill Birkbeck College, University of London Agree Confident
Falling unemployment rates, a growing proportion of people employed is still going to put some upward pressure on wages, but it is not clear how much, or at what level of unemployment it really starts to have a noticeable effect. The quote from Mario Draghi above seems right: "As the labour market tightens and uncertainty falls, the relationship between slack and wage growth should begin reasserting itself. But we have to remain patient."
Wouter Den Haan London School of Economics Agree Confident
Traditionally that has been the case and there are good theoretical reasons for this. If labour markets are truly "strong," then this should lead to higher wages and that will have to lead to higher prices at some point. A low unemployment rate may not be sufficient to talk about a "strong" labour market.
Ethan Ilzetzki London School of Economics Agree Confident
It is true that the (unconditional) correlation between inflation and employment has weakened, but I have not yet seen persuasive evidence that this implies a structural flattening of the Phillips Curve, nor a good explanation for why this may have happened. I therefore trust basic price theory, which suggests that wages (and therefore prices) will eventually increase if labour market tightness persists. The Phillips Curve was never a causal relationship, but rather a correlation that should hold empirically as long as the majority of price variance is due variation in aggregate demand. When aggregate supply shocks are dominant (as in the 1970s) this relationship reverses. Alternatively, if central banks successfully target inflation and/or inflation expectations are strongly anchored, we’d expect to see a zero correlation between inflation and employment, as is the case today.
Sean Holly Cambridge University Neither agree nor disagree Confident
Angus Armstrong National Institute of Economic and Social Research Agree Confident
I believe that a strong labour market is an indicator of inflationary pressures. However, the relationship is much looser than 'output gap' analysis would suggest. Global integration has had an influence on domestic wages, particularly at the lower end of the spectrum through some extent of factor price equalization. This has weakened the trade-off between wage and inflation. Ignoring international aspects has led to misreading the extent of excess or shortage of demand.
Sir Christopher... London School of Economics Disagree Confident
Because of structural changes in the labour market such as deunionisation, shift to services, foreign competition, automation and the rise in inequality, it is not clear any more what is a strong labour market. If by it we mean low unemployment then it is not a good indicator of inflationary pressures because in the old days the inflationary pressures originated in manufacturing and unionised workers which have become too small a group to matter
Etienne Wasmer Sciences Po, Paris Disagree Very confident
It seems that changes in unemployment in a country tell is very little on the reserve/shortage of labor available to firms in this country: there are large participation margins and many countries in the world who can supply labor. If local wages are too high, it is now very easy to subcontract some of the activities to other countries.
Michael McMahon University of Oxford Neither agree nor disagree Confident
I see merit in the arguments for why the Philips curve is hiding. With that in mind, at a time of a weakened relationship, then the extent of the pass through from labour market strength to future inflation will be reduced and with it the quality of indicator that labour market variables play in forecasting inflation.
Martin Ellison University of Oxford Agree Confident
I'm concerned that our ability to accurately measure inflation is deteriorating over time. The statistical authorities find it difficult to measure inflation when the nominal recorded cost of many services is zero (Skype, Facebook etc), so the strong labour market may be a good indicator that the economy may be overheating even if inflation is muted. Strong labour market performance is an indicator of *something* and that something may call for policy actions.
Simon Wren-Lewis University of Oxford Disagree Confident
There are many reasons why the NAIRU may have fallen over the last decade. One that is seldom talked about is productivity. Productivity growth in many countries has been low since the financial crisis. However it seems unlikely that this reflects an equivalent decline in fundamental technical progress. Instead what seems more likely is that many firms have put off investment in labour saving improvements because of weak growth. If labour does become scarce, we are more likely to see a catch up in productivity growth than rising inflation.
Ray Barrell Brunel University London Agree Confident
Low unemployment along with reasonable employment growth give a good indication that the labour market is strong, and that inflation might rise. We are in a position where unemployment looks low, and inflation is above target. Hence a rise in interest rates could be justified. However, labour markets are more complex than the poor description of them given by the Phillips Curve. The ‘Curve’ is at best a statistical description of a reduced form relationship and at worst an empirical mirage. Even if we assume that there is at any point in time a unique equilibrium unemployment rate, we do not know what it is, but we do know it changes all the time. Equilibrium unemployment will depend upon the factors affecting the demand for and supply of labour, as well as on the direct and indirect tax rates and the real exchange rate, and these are not constant. Even if they were unchanging, technical progress and changes in average labour quality determine equilibrium real wage growth, and these also change over time with rates of innovation, migration and other factors. We work in the dark, but careful analysis of data can help us find our way. It would all be easy if we knew what equilibrium unemployment currently is, but we do not. If unemployment is below the equilibrium then wages should rise more than would be anticipated, and this would lead to increases in costs and hence rises in prices. This would be a good indicator of potential inflationary pressures, and monetary policy could mechanically respond. We are not in that world, and probably never have been. Judgements have to be made, but they must be based on more information than can be summarised in a shifting and unstable Phillips Curve. The ‘Curve’ is a useful teaching tool, and of value for commentators, but of little use for serious policy decisions.
David Cobham Heriot Watt University Agree Very confident
In principle, yes. But we need to revise our ideas of 'strong' so as to take account of the innovative devices used by employers (zero contract hours, tightened work discipline, that sort of thing) as well as the much greater pressure from job centres and the benefit system on the unemployed to take poor jobs: the older period search models which can be seen as lying behind the Jackman, Layard and Nickell conceptualisation of the Phillips relationship are no longer adequate (but need to be supplemented rather than thrown away). Such a reappraisal which focuses on long term changes in the balance of power in the labour market, as well as cyclical variations, would also illuminate current issues regarding inequality.
Akos Valentinyi University of Manchester Agree Very confident
Tony Yates University of Birmingham Neither agree nor disagree Confident
It depends on what is causing the strong labour market. Positive supply side shocks can mean strong labour market quantities without inflationary pressure. The Phillips curve is really just a bivariate correlation. The modern equivalent embedded say in a many variable DSGE model can quite easily generate fluctuating Philips Correlations, as the mix of shocks changes.
Richard Portes London Business School and CEPR Neither agree nor disagree Not confident
Joseph Pearlman City University London Disagree Very confident
Trade unions now have a far smaller membership than 40 years ago, making it much more difficult for them to influence wages as significantly as in the pre-Thatcher era. I am not convinced about the influence of the global marketplace; we are only now back again at the same level of trade/GDP as we were in 1914, and that did not then prevent trade unions putting upward pressure on wages. Evidence from migration research indicates little downward effect on wages apart from at the very lowest incomes. Central banks may anchor expectations of inflation but if inflation goes up significantly, then wages will follow. So the emasculation of trade unions is what I believe counts the most.
Fabrizio Coricelli Paris School of Economics Disagree Confident
Philippe Martin Sciences Po, Paris Agree Very confident
The answer depends how one defines strong labour market. Standard unemployment rates may not be enough to judge the strength of the labor market. Participation rates, non standard jobs, part time jobs... may be as or more important.
Lucio Sarno Cass Business School Agree Very confident
Jean Imbs Paris School of Economics Neither agree nor disagree Extremely confident
Panicos Demetriades University of Leicester Disagree Very confident
The Phillips curve has indeed weakened for the valid reasons you have provided above. The most important ones are the increased labour mobility and decreased union power, These reasons aren’t transient, they are here to stay.
Pietro Reichlin Università LUISS G. Carli Disagree Not confident
I tend to agree with Summers that changing conditions in the labor market (de-unionization, digitalization, globalization, etc.) may have contributed to the weaker negative correlation between inflation and labor market slackness.
Jim Malley University of Glasgow Agree Very confident
Patrick Minford Cardiff Business School Agree Confident
The Phillips Curve error term is sensitive to the factors listed above: immigration, global competition and so on. Plainly these have been dampening the response to unemployment; however there is no reason to believe this response has gone away.
Alan Sutherland University of St. Andrews Agree Confident
Kate Barker British Coal Staff Superannuation Scheme Strongly Agree Very confident
yes - it's hard to think that if people can get jobs and are able to pick and choose a little - this won't put upward pressure on wages, ceteris paribus. This could of course be offset if it prompts a productivity response.
Jagjit Chadha National Institute of Economic and Social Research Agree Confident
I shall interpret strong here as tight, in the sense that we are then close to or approaching some notional level of full employment. The most significant part of most firms' costs are connected with labour inputs and a significant fraction of output is still related to what used to be called the non-tradable sector. These two factors will tend to imply that a tight labour market represents a risk to a given inflation outlook but at best the labour market alone, is a noisy indicator of overall inflationary pressures.
Evi Pappa European University institute Disagree Confident
David Miles Imperial College Agree Confident
The real issue is what does "strong labour market" mean. Unless you give that concept some empirical content the proposition is empty. And it is very clear that assessing "strong" by reference to a some measure of unemployment is tricky when the nature of jobs has changed as has unionisation and the openness of economies to flows of goods, services and people.
Fabio Canova BI Norwegian School of Management Agree Confident
it is necessary but not sufficient
Paul De Grauwe London School of Economics Agree Confident
Costas Milas University of Liverpool Neither agree nor disagree Confident
The relationship between inflation and unemployment is by no means linear and depends on the starting point of inflation. In a brand new paper joint with my colleague Michael Ellington (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2882494) we show that UK inflation is driven by monetary developments when inflation is already high (above 3%). Therefore, at high inflation rates, MPC should pay closer attention to money growth rather than labour market developments.
Robert Kollmann Université Libre de Bruxelles Neither agree nor disagree Confident
It all depends on the underlying shocks. When aggregate demand shocks are driving the cycle, then strong labor markets indicate inflationary pressure. But not when aggregate supply shocks are dominant.
Ricardo Reis London School of Economics and Columbia University Strongly Agree Very confident
As a simple plot, or correlation, the Phillips curve is a very unreliable economic law. But, as a structural relation that encapsulates the strength of nominal rigidities, monetary non-neutrality, or deviations from the classical dichotomy, it is one of the most powerful, effective, and useful pieces of applied economics. I am interpreting the vague term "strong labor market" as a measure of real activity standing fot the structural interpretation.
Jonathan Temple University of Bristol Agree Not confident
Sylvester Eijffinger CentER, Tilburg University Strongly Agree Very confident
There is no substantial empirical evidence that the Phillips curve relationship has weakened permanently not that a strong labour market shouldn’t be a good indicator of building inflationary pressure. Mario Draghi is right that we must be careful with interpreting inflationary expectations.
Roger Farmer University of Warwick Disagree Very confident
The Phillips Curve has not been a good characterization of data in any advanced economy since Phillips published his paper on the relationship between wage inflation and unemployment in the UK in 1958. Subsequent empirical developments that purport to find a Phillips Curve in data are not estimating the same relationship uncovered by Phillips, They are instead estimating the “expectations augmented” Phillips Curve. This is an irrefuatable theory that contains an unmeasurable concept; the average subjective expectations of markets participants. The Phillips Curve was a reduced form relationship that existed during a period when world monetary arrangements were governed by the Gold Standard. There is no reason to expect the same reduced form relationship to hold in a world of fiat monies. In my view, central banks and most practicing macroeconomists are currently working with a flawed theory.
Roel Beetsma University of Amsterdam Neither agree nor disagree Not confident
I think there are transition shifts (such as inflow of new workers, intensifying global competition) hiding the effects of falling unemployment rate on price and wage pressures.
Omar Licandro University of Nottingham Disagree Very confident
Jan Eeckhout University College London Agree Very confident
Yes, but less so now that decades ago.
Cédric Tille The Graduate Institute, Geneva Agree Confident
Michael Wickens Cardiff Business School & University of York Disagree Very confident
For many years the various versions of the Phillips curve have failed to capture a significant relation between inflation and unemployment, or even between wage inflation and labour market pressure. It is, and has been for many years, one of the weakest links in the New Keynesian models of the UK and US. It is interesting that the wage-price sector of the Smets-Wouters model is rejected even though this is one of the main things that they thought would result in an improvement over the price flexibility of New Classical models. A possible reason in the UK is that the relation is based primarily on demand effects whereas in recent years the labour market has been dominated by labour supply effects, namely, a large increase in the supply of (mainly unskilled) labour.
Christopher Martin University of Bath Agree Not confident
In recent decades, a tight labour market - measured by the number of vacancies per unemployed worker- has been a reliable indicator of inflationary pressures The mechanism is simple: firms needs to offer higher wages to recruit workers, this increases costs and feeds through into higher prices. There is no reason to think that this mechanism has broken down. But the working of the mechanism may well have changed and that the dismal decade since the financial crisis has eroded some existing labour market linkages and led to the growth of new connections. And the equally dismal prospect of Brexit, with possible dislocation of inflows fo workers to the UK, is probably already changing the labour market. So while a strong labour market stills leads to inflationary pressures, we need to to re-assess how to measure labiur market strength. Evidence from the US suggests that the "vacancy yield", the number of hires per vacancy has fallen. There is some evidence that this is because firms have gained power in the labour market over the past decade and are using this to suppress wages, particularly for job-to-job movers. It is not clear how much of this applies in the UK- where our data is not so detailed - but the US experience has usually been a good guide to the UK, and the argument that firms have gained power in the labour market fits in with anecdotal evidence. The US evidence suggests that one indicator may have kept its power as an inflationary canary: the number of workers hired from unemployment. It would be sensible to keep an eye on this in the UK
Jürgen von Hagen Universität Bonn Disagree Very confident
Deunionization, globalization, and migration are no new phenomena. I am more skeptical about central banks' ability to anchor inflation expectations. They promote that view, but, in fact, their anchor - the inflation target - has been much higher than actual inflation for many years now, suggesting that the anchor does not have much credibility.
Harris Dellas University of Bern Strongly Disagree Very confident
There exists no PC
Stefan Gerlach EFG Bank Agree Confident
I believe that the relationship between the state of the labour market and inflation remains unchanged; what has changed is how best to measure the strength of the labour market. In the past when part-time work was limited, the level of unemployment worked well. These days, when many firms want to have some part-time staff and many workers prefer part-time employment, other measures, such as the number of workers that are involuntarily part-time employed or working shorter hours than they would like, might work better.
Antonio Fatás INSEAD, Singapore Agree Confident
Francesco Giavazzi IGIER, Università Bocconi, Milano Agree Confident
the Phillips Curve is still alive
Philip Jung University of Dortmund Disagree Confident
Gino A. Gancia CREI and Universitat Pompeu Fabra Agree Not confident

Monetary Policy and the labour market

Participant Answer Confidence level Comment
Gianluca Benigno London School of Economics Agree Confident
The signs of wage inflation are not so worrying to demand an immediate response from a monetary policy point of view. There is room to wait to be certain that the real wage developments are indeed consistent. The problem in shifting the policy stance (more than the increasing of 0.25% in the nominal interest rate per se) is that it has lead to an appreciation of the British pound which eventually will reduce CPI inflation and limit the needed rebalancing of the UK economy. Compared to about 1 year ago the USD/GBP exchange rate has appreciated around 16%. Monetary policy is not neutral with respect to productivity developments as I discuss in a recent vox column (https://voxeu.org/article/keynesian-model-long-run-growth) and in a small open economy this link is further amplified by exchange rate changes. The approach highlighted in the vox column requires a different way of thinking about the link between cyclical fluctuations and trend. This approach would then suggest that waiting for a sustained pickup in business investment that would result in higher productivity growth would minimize the risk of a policy mistake and could have long lasting effect. Moreover trying to mimic the normalization of monetary policy as in the US economy does not keep into account a key difference in terms of fiscal stances (i.e. expansionary in the US, neutral at best in the UK). If anything with uncertainty related to Brexit, monetary policy should tolerate inflationary pressure along with a more depreciated exchange rate. Achieving that required a change in the stance of monetary policy.
Nezih Guner CEMFI Strongly Agree Confident
The key challenge for the policy makers is to judge the cost of raising interest prematurely rates in an environment with growing income inequality, income uncertainty and populism. I do not think we have good models to judge this trade-off, in particular how an economic slowdown can fuel populist sentiments with potentially important long-run consequences.
Ramon Marimon European University institute and UPF-BarcelonaGSE Agree Confident
From the perspective of the labour market there is no reason to increase interest rates, from the perspective of still indebted economies in need to rebuild their capital even less. Nevertheless, from the international perspective of capital markets one also needs to take into account what others do.
Richard Dennis University of Glasgow Agree Very confident
Fabien Postel-Vinay University College London Agree Not confident
John Driffill Birkbeck College, University of London Agree Confident
Real wages are surely an important piece of information, but so are money wages and prices, and other indicators of incipient inflation, and indicators of robust growth continuing in the coming months and years.
Wouter Den Haan London School of Economics Strongly Agree Confident
yes, central banks have built enough credibility and inflation expectations seem well anchored. So it makes sense to be cautious and wait for convincing evidence of significant wage and inflation pressure.
Ethan Ilzetzki London School of Economics Disagree Not confident
The Bank of England has an inflation target. Inflationary pressures are starting to mount and it is the Bank's responsibility to respond to them. I agree, however, that the outlook is sufficiently uncertain that taking a gradual approach to interest rate increases is advisable.
Sean Holly Cambridge University Disagree Very confident
Angus Armstrong National Institute of Economic and Social Research Disagree Confident
The issue is less about raising interest rates at all, more about whether the adjustment should be gradual. On this point I would support the gradual adjustment. But there are other signs of reaching full capacity than just past real wage growth. For example, increasing leverage is related to risk and the cost of capital. In my view policy makers should take a broader interpretation of their mandate and include other signs of being somewhere near to full capacity.
Sir Christopher... London School of Economics Agree Very confident
Given the uncertainties that we are facing in labour markets, the signal given to markets by an MPC interest rate rise will confuse markets even more. Better wait until there is a clear understanding of what is going on
Etienne Wasmer Sciences Po, Paris Disagree Confident
Slowly raising interest rates is becoming urgent ; not because of the inflation of commodity ; but because of inflation of financial and real assets such as housing. I would like to see a progressive rise in interest rates in order to reduce at a slow pace the price of these inflated assets.
Michael McMahon University of Oxford Strongly Agree Confident
I am normally quite hawkish in taking steps against inflation. However, I believe that today there are reasons for waiting. First, the labour market data has, in recent years, suggested building inflationary pressures that turned out to be false dawns. Second, there remains huge economic uncertainty going forward in the UK. Finally, given the erosion of real wages in the aftermath of the financial crisis (still 5% below 2008 levels in recently-released data) and the fact that real wages have not grown for 2 years (April 2018 ONS release), I believe that the costs of falling behind the curve are not huge.
Martin Ellison University of Oxford Agree Confident
The downside to inaction is "letting the cat out of the bag", i.e. setting in motion a chain of events that will be difficult to control. In the current climate I don't see what this could be. If inflation were to rise then the central bank has plenty of ammunition in terms of interest rate hikes or QE reversals, so it seems premature to act now.
Simon Wren-Lewis University of Oxford Strongly Agree Extremely confident
There are many reasons for this, including the point made in answering the previous question. Most important, however, is the asymmetry of costs if policy is wrong. Tighten too late and we get a modest inflation overshooting. Tighten too early and, because inflation is sticky near zero, it may be years before central banks realise their mistake, leading to very large welfare losses.
Ray Barrell Brunel University London Disagree Confident
All periods over the last thirty years appear to have involved great uncertainty, and today looks less uncertain than, say 2008-9, and no worse than average. It just happens to be today, so people worry about it. After nearly ten years of weak wage growth it would be a good bet that this will continue, and policy should be based on this assumption. Policy can be flexible if this is not the case. The labour market might change, and weak technical change and slow growth in labour input quality may suddenly switch to higher growth. But then, both could also slow even further. Risks are not always symmetrical, and higher rather than lower growth of earnings may be more likely, but the most likely outcome is that this year will be like the last five years. Uncertainty is not a reason for inaction. Policy makers could afford to wait in raising interest rates, but it would be best if they raised rates now. Small changes in interest rates have very little impact on either output or inflation. A delay may mean larger increases in future for the same inflationary pressures, and these larger, delayed movements may involve more costs than several, more timely, small increases in rates. The costs of acting too early are small, coming mainly through the exchange rate. Interest rates should start to rise now and rise twice more this year.
David Cobham Heriot Watt University Strongly Agree Very confident
Yes, the MPC should wait: it is not yet clear that real wages are rising in any significant way, particularly if account is taken of the continuing high level of underemployment as highlight in the article by Davids Bell and Blanchflower in the latest National Institute Economic Review.
John VanReenen London School of Economics Agree Very confident
Akos Valentinyi University of Manchester Disagree Very confident
Tony Yates University of Birmingham Agree Confident
I think the costs of being wrong about this and realising that inflationary pressure was in fact building and monetary policy was too weak are less than the costs of erring on the side of tighter policy. The latter could mean a greatly prolonged period back at the effective lower bound to interest rates.
Richard Portes London Business School and CEPR Strongly Agree Confident
Joseph Pearlman City University London Agree Not confident
Although I agree that one should wait for more certainty about wage growth before raising interest rates, I am not convinced that this is the best thing for the country. With TFP growth particularly low, it is highly likely that we need poorly performing firms to drop out of the market, and they will only do so once their rates of return are below the interest rate. So it is possible that an increase in interest rates would benefit the country even if it led to a temporary loss of jobs.
Fabrizio Coricelli Paris School of Economics Disagree Confident
Credibility is an issue. Moreover, impact of low rates on financial markets cannot be neglected.
Philippe Martin Sciences Po, Paris Strongly Agree Very confident
The uncertainty on the level of output gaps and potential output is very large and recent research has shown that standard measures of potential output responds to monetary shocks. In this type of situation better to err in the direction of waiting observed pressures on wages and prices to tighten.
Jordi Galí CREI, Universitat Pompeu Fabra and Barcelona GSE Agree Confident
But it would be a mistake to wait too long and be forced to increase rates too rapidly.
Lucio Sarno Cass Business School Disagree Extremely confident
Jean Imbs Paris School of Economics Strongly Agree Very confident
Panicos Demetriades University of Leicester Strongly Agree Extremely confident
This is not the time to be raising interest rates, the BoE should wait to see how Brexit negotiations unfold. Let’s not forget any inflationary pressures to date have been due to the falling value of the pound because of Brexit uncertainty. If we have a positive outcome, inflationary pressures will recede further.
Pietro Reichlin Università LUISS G. Carli Agree Confident
In addition to wage growth being less responsive to falling unemployment, I would add that credit growth is not particularly strong and the overall economic scenario appears to be subject to many risks.
Jim Malley University of Glasgow Agree Confident
Patrick Minford Cardiff Business School Strongly Disagree Extremely confident
The UK economy is no longer in 'crisis' mode, justifying emergency low interest rates close to the zero bound, and also bank reserves, from QE operations, of around 25% of GDP. This situation creates large distortions in savings markets: on the one hand absurd incentives to lend on car loans and mortgages, while on the other denying credit to small businesses (which are risk-penalised). Ideally raising rates and reducing QE would be accompanied by an easing of bank regulation and 'macro-prudential controls'.
Alan Sutherland University of St. Andrews Agree Confident
Kate Barker British Coal Staff Superannuation Scheme Disagree Confident
Wages are not the only thing that matters. we learned about the dangers from credit imbalances in the 2000s. Monetary policymakers should worry about the risks of economic volatility as well as just the inflation target. This is all supposed to be taken care of by the FPC - but unlike Ben Broadbent my view is that the FPC and MPC should be one - though I have not always thought that.
Jagjit Chadha National Institute of Economic and Social Research Disagree Confident
The critical point here is that raising rates from the near zero does not imply tight monetary policy. Rates could be raised a number of times in small increments and still be providing a monetary stimulus. Whilst all measure of economy-wide slack are also uncertain, to the extent that low productivity trends have crimped supply, there is a danger that even small increases in unit labour costs may pose an inflation risk. The question is more whether we should continue to wait for a normalisation in rates. Some of the risks of expectations of significantly higher interest rates can be offset with central bank projections of Bank Rate.
Evi Pappa European University institute Agree Confident
David Miles Imperial College Disagree Not confident
It is not likely that waiting until it is clear that a change in rates is needed is optimal. Fortunately, rate changes can be reversed, which is one reason by why waiting is a mistake.
Fabio Canova BI Norwegian School of Management Agree Very confident
Albert Marcet Institut d’Analisi Economica, CSIC Agree Confident
Paul De Grauwe London School of Economics Agree Confident
Costas Milas University of Liverpool Agree Confident
In the presence of elevated economic uncertainty, monetary policy tightening becomes less effective (see https://www.sciencedirect.com/science/article/pii/S0261560617300943). The reason is that elevated uncertainty motivates agents to postpone decisions until more precise information becomes available, and this cautiousness makes them less responsive to changes in the economic environment, including the interest rate. The implication is that an additional interest rate hike will have a much smaller impact on inflation and GDP growth than conventional wisdom suggests. Is it not better to wait until we have a clearer picture about the economy? Luckily enough, the GDP reading for 2018Q1 will come before the May interest rate decision so MPC members will have a clearer picture of the economic outlook before pressing the button.
Robert Kollmann Université Libre de Bruxelles Agree Confident
Ricardo Reis London School of Economics and Columbia University Agree Confident
Jonathan Temple University of Bristol Agree Not confident
Kevin Hjortshøj... Oxford University Agree Confident
Sylvester Eijffinger CentER, Tilburg University Strongly Agree Very confident
Monetary policy makers should be careful with interpreting inflationary expectations but must also be aware of large overliquidity they have created by their expansionary monetary policies after the financial crisis and react quickly when real wage developments are going out of hand.
Roger Farmer University of Warwick Neither agree nor disagree Confident
I am in favour of continuing with a gradual rate tightening cycle, whilst simultaneously offsetting potential harmful effects on the asset markets through active macro prudential policies. In the current environment, the Bank pays interest on reserves. It is operating a ‘floor system’. Raising the rate on reserves at the same time as raising the overnight lending rate will not lead to monetary tightening in the same way as it would in a world where these rates were allowed to diverge: (a corridor system). Responsibility for maintaining ‘strong sustainable balanced growth’ should be transferred to the Financial Policy Committee, strengthened with the tools to conduct more aggressive interventions in the asset markets.
Roel Beetsma University of Amsterdam Disagree Not confident
I am not too familiar with the specific situation in the UK, but if I understand correctly, there are signals about increasing inflationary pressures in particular increasing costs of imports, while at the same time monetary policy changes take about one-and-a half years to feed through into real activity.
Omar Licandro University of Nottingham Agree Very confident
Gernot Müller Eberhard-Karls-Universität Tübingen Agree Confident
Jan Eeckhout University College London Agree Confident
Alexander Ludwig Goethe University Strongly Agree Very confident
Cédric Tille The Graduate Institute, Geneva Strongly Agree Very confident
Michael Wickens Cardiff Business School & University of York Disagree Confident
Although unemployment is little related to inflation, wage growth does contribute to inflation. But as wage growth is not the only, or recently even the main factor, affecting inflation, the monetary policy decision should not focus just on wage growth. As higher inflation will result in higher wage growth there may even be reverse causation. This suggests a more aggressive anti-inflation policy now rather than wait for wage growth as this will dampen any transmission from wage growth.
Christopher Martin University of Bath Agree Confident
It seems pretty clear that policymakers could afford to wait before raising interest rates. The labour market seems muted, the effects of the large devaluation are working their way out of the inflation figures without become embedded as wage pressure and the rate of domestic growth remains weak. But could is not the same as should. Although interest rates are still extraordinarily low by historical standards, there is no evidence that firms are using this to finance a wave of new investment. Tentative evidence of a pick-up in productivity growth may suggest that the issue of "zombie firms" was overstated. So the arguments against raising rates back towards more normal levels are not as strong as they were. On balance, it might be sensible for the MPC to continue its slow and cautious path of interest rate rises.
Jürgen von Hagen Universität Bonn Agree Confident
Harris Dellas University of Bern Agree Confident
Stefan Gerlach EFG Bank Disagree Confident
Monetary policy is exceptionally expansionary in many countries, which has been appropriate given economic conditions. But they have a long way to go before interest rates have been "normalised" and there is a risk that they might fall behind the curve. That suggests to me that a little bias in favour of higher rates might be appropriate.
Antonio Fatás INSEAD, Singapore Agree Confident
Gino A. Gancia CREI and Universitat Pompeu Fabra Agree Confident
Philip Jung University of Dortmund Disagree Confident