Our work is only possible through the generosity of private philanthropy. Find out how you can support our mission and can contribute to our work.
Join the HJS mailing list and keep up to date.
Chair: Alan Mendoza, Director of The Henry Jackson Society
Host: Gisela Stuart
Speaker: Kirsten Forbes, Professor
Tuesday 26th January, 13:00-14:00
Committee Room 18, House of Commons, SW1A 0AA
Transcript written by Rachel Oliver
Right, ladies and gentlemen. We’re going to kick off. Gisela Stuart is attending a question in the House of Commons. In fact, you can see it happening right now, where she has to make an intervention. So she will be with us as soon as that happens. However I am pleased that Professor Kristin Forbes is with us. Because without Kristen, obviously, the event would not go ahead. So we have the rather charmingly and interestingly named, “Tale of Two Labour Markets: The UK and the US” which she is going to take us through. She has a very distinguished record and pedigree in this department. Apparently she is Monetary Policy Committee member for the Bank of England, where she has been since 2014. She is also a professor at the M.I.T., where she has been for a long time. And it is such a long title I’m not going to read it out, but suffice to say she’s very clever. But she’s also had this history, I think, of double-hatting. Doing public policy, as you’re doing at the moment, but also being an academic. And you’ve been a deputy Assistant Secretary at The Treasury, the Council of Experts at the White House, and indeed more locally in Massachusetts, quite a long-term advisor to the economic policy there. So we’re delighted you’ve joined us today. The format is that Kristin will give her presentation, we’ll then open up to questions and conclude just after 2pm. Kristin, over to you.
Thank you very much. It’s a pleasure to be here today. Especially for a group that honors the long-standing historic relationship between the USA and the UK – a special relationship that will be the focus of my comments today. Before I get to that, I’m going to start with another special historic relationship, honoured by Dickens in a Tale of Two Cities. The relationship between the United Kingdom and France. I’m sure you are all very familiar with the opening quote of a Tale of Two cities by Dickens, so I will not read it to you. Part of, I’m sure, your proper British education. But what is striking is, if you read this quote today, how well it applied to the UK labour market today. For example, in many ways the UK labour market today could be described as “the best of times”. The unemployment rate is at 5.1% today. That’s the lowest rate in over 10 years. Over 260,000 net new jobs have been created over the last three months in the UK. So that now there’s over 1.7 million new jobs than before the financial crisis. The unemployment rate for the working age population in the UK is 74%, the highest since records began in 1971. But, by other measures, the UK labour market could also be described as the worst of times. Wage growth has slowed recently, with whole economy total payrolls slipping from 3.3% through May, to 2.0% in the three months through November. Median incomes in the UK have only grown by 1.6% over the last 4 years, so that after adjusting for inflation, median household income is still about 2% below its peak in 2009/2010. So many workers who have found employment recently may feel Dickens’s “spring of hope”, but if wages do not start to go up soon, they may slip into his “winter of despair”. So this dichotomy in the labour market is going to be the key theme of my comments today, and it creates some substantial challenges for monetary policy. Let me just summarise some of the challenges, and then I’ll go through in more detail. On one hand, our labour market quantity measures suggest a labour market that is very healthy, very near equilibrium. If the challenges companies are currently facing to find employees continue, and they accelerate, this could drive a sharp pick up in wages, a sharp pickup in inflation that would suggest we already may be behind the curve. We should be tightening monetary policy now. But on the other hand, if you look at the nominal measures – wages and prices in the UK – these are quite a bit more muted. Suggests that there may be more slack in the labour market. Firms are not under so much pressure that they need to pay higher wages, and if this slow growth continues and continues to feed through into low inflation, currently at 0.2%, combined with downward pressure on prices from cheap energy and Sterling’s appreciation, there’s little risk of wages spiking or inflation spiking. In which case, monetary policy could be on hold and there’s no reason to tighten monetary policy. So correctly interpreting exactly what is going on in the labour market, which of these factors is the true underlying strength of the labour market, is going to be critically important in setting monetary policy. If we misinterpret this, we could fall into what Dickens calls an “age of foolishness”. But my goal today would be to steer us in the other direction towards Dickens’s “age of wisdom”. Now to do that, I’m going to take a close look at the UK and US labour markets, and I’m going to do, basically, three things today. First, talk about similarities between the UK and US labour markets. And in particular, comparing how the UK is compared to a country that has just started tightening monetary policy, the US. Then I’ll focus on four key differences between the UK and US labour markets. More specifically, I’ll focus on differences in labour force participation, on self-employment, on migration, and the consistency and stability of the recovery. And third, I’ll talk about what all this implies for wage growth and inflation in the US and UK. And finally, I’ll conclude by trying to tie it all together for what it implies for monetary policy in the United Kingdom. And welcome. So that is the game plan, let’s dive right in with similarities in the UK and US labour markets. Now I’m going to show you a lot of graphs and a lot of numbers, so hopefully you’re ready. But one thing that will stand out is how similar the UK and US labour markets are, in fact when you look at data in graphs sometimes it’s very difficult to differentiate which is the UK and which is the US. So just to keep you on your toes, what I’m going to do is disguise the UK and the US labour markets as two cities. I’m going to channel Dickens here. And I’ve picked two cities to hide their identities. Two cities both in the US and in the UK, two lovely cities – Dover and Exeter. Obviously you all know Dover, for its famous white cliffs, and Exeter, with its well-known university. Those are also two cities in New Hampshire, which is a state where I grew up and spent a lot of my time. No gorgeous white cliffs in Dover in New Hampshire, but here are some stunning waterfalls which look somewhat similar from far away in the back. And Exeter is a city also well-known for its wonderful institution of learning. So what I’m going to show you in the next series of graphs is show you the UK and the US, but hidden as Dover and Exeter. I will warn you, in a few minutes I’m going to take a poll and see who can guess. Does Dover represent the UK or the US, or is Exeter the UK or the US? We’ll see how well you know the little details in the data for the UK at least.
So with that, let me just dive right in. what are some similarities in the broader macroeconomy relevant to the labour market, and the specific labour markets of the UK and the US. First similarity – what’s been happening to output growth? Both countries Dover and Exeter, or cities, saw output sharply fall during the crisis. Then it recovered and bounced back, sort of bounced around for the initial stage of the recovery. Very strong growth in 2013/2014. And since then, growth has moderated in both countries. So that as of Q3 of 2015, both had GDP growth of around 2%.
Interest rates, policy rates in both countries, fell sharply after the crisis and then stayed at close to zero for a number of years after the crisis. You will note I stopped the graph in November because the US did raise interest rates, and that would be a dead giveaway which was the UK and the US. But at least until November, both countries had kept interest rates near zero for an extended period to support the recoveries.
Another set of similarities – what’s happening to the exchange rates? Or what has happened to the exchange rate with inflationary pressures? Both countries have had very sharp exchange rates appreciations, from the spring of 2013 to its peak of about 22-25%. Still, it has since depreciated, so now it’s a bit weaker. But this peak appreciation is still weighing on costs for both countries, reducing important prices, driving down inflation. So when this effect – this sharp currency appreciation is combined with sharper falls in energy prices and food prices, it’s meant very low inflation in both countries. I didn’t hide the identities here, as this might be a giveaway, but as you can see, headline inflation and core inflation in the two countries is basically identical.
Moving to the labour market, the similarities are even stronger. Here are unemployment rates in Dover and Exeter, the UK and the US, sharp – very remarkable decline in the unemployment rate in both countries after the crisis. So that right now, unemployment in both countries is at about around 5%, which is believe to be around the equilibrium unemployment rate for both countries. What that means is that when unemployment rates get to about the equilibrium rate – that’s when wages generally start to take off and you start to see wage growth accelerating, often quite quickly. In both countries, at just about that point. The challenge though is, you see these dotted lines here, the dotted lines are pre-crisis averages for where unemployment was in Dover and Exeter before the crisis from 1997 to 2007. And you can see according to one country, unemployment is below its pre-crisis average, another is getting to be right about its pre-crisis average. And we often look at how these statistics are doing, relative to equilibrium or pre-crisis averages, to know really how tight the labour market is. But one challenge with that is – what is equilibrium? What is normal? What is the comparison we should be using? You might say 1997 – 2007 isn’t a great period, maybe the labour market was too tight then, so that’s not a good comparison. Or you might say there have been some very serious structural changes in the economy, so what was equilibrium that – if we knew what it was then – that may not be the appropriate equilibrium rate right now. So measuring what’s equilibrium is very hard for a number of these labour market measures. So what I also like to look at is measures of labour market churn. How efficient is the labour market? How quickly are people able to find jobs? How many vacancies are listed by companies? Are people comfortable switching around jobs? That is not what I want to look at. The Bank of England wants to pop in there. Okay well I’ll talk through the next few slides.
If you could see them, the next few slides would have shown measures of labour market churn. Basically, how willing people are to leave their jobs, how willing they are to quit, how easy it is to move between jobs, and how many vacancies are listed by companies. And what you would see, if the slides were working, would be basically identical lines for the UK and the US. Both the countries have seen big improvements in people willing to quit their jobs, move to another job in another company, big increase in vacancies listed by companies. Companies are now listing more vacancies than before the crisis. We have record levels of vacancies listed by companies, showing that companies have a lot of empty slots they are trying to fill and are unable to fill. The only measures of this labour market churn that is not quite at equilibrium levels here yet, is quits. There does seem to be some hesitancy of workers, especially in the country I labelled Dover, to quit. So not quite as much confidence in the recovery. But otherwise, these measures of labour market churn show very, very tight labour markets. So the bottom line, is, if I got all of those graphs up, would be to show you that labour markets are very tight in the united kingdom. Most measures of labour market churn – most of these measures that I would show you, unemployment rates, short-term unemployment rates, long-term unemployment rates, people’s willingness to move around between companies, companies trying to hire, very tight – not a lot of slack in the labour market. The one area where there is, though, some questions – although still similar between the two countries – is earnings. Earnings are not going up as quickly as they normally would be doing when you have a labour market that is this tight. There is, for example, this shows median earnings growth rate – well below where they were below the crisis. Median earnings have been somewhere between two to three percent. Very different from the message you get about how tight the labour market is, how low unemployment rates are, how comfortable workers are moving between jobs. So this has led to some questions of maybe the labour market is not that tight or maybe we are just looking at the wrong measures of wages and this is where the story does get more confusing.
Different measures of wages do send some wrong signals. For example if you look at wages adjusted on how many hours people are working, you see that wages are going up quite a bit in one country, quite a bit higher in Dover according to this graph. Wages are still below pre crisis averages, these dashed lines, but it does show some more strength. Some of you might actually be looking at the chart and saying I figured out which is Dover and which is Exeter. You might be thinking well the red line, Dover and that must of course be the US because the US have raised interest rates so they must have higher wages. But some of you might also be thinking o no wages went up in the UK so maybe Dover is the UK. Not quite so easy.
Another graph just to show you how confusing it can be looking at this wage data. This is real wage growth. This is wage growth adjusted for inflation. Inflation is fit very well so were thinking that our companies may not need to pay as much as a wage gain because inflation is so low. Individuals may be earning more when you adjust for the low cost of the items they buy. If you make this adjustment, you see wage growth is actually been quite strong in one country, it is now above pre crisis averages from the most recent data, whereas wage growth has been more moderate in the other country.
So the bottom line is wage data overall does show some weakness and loss of momentum, especially considering how tight labour markets are. The different measures do send some very different signals. Different measures do show differences in terms of which country is ahead in wage growth relative to the other. So all in all, summarising all that, the graphs which you could see show recovery of GDP in both countries, UK and US or Dover and Exeter have solid growth rates right now. Policy rates are still near zero since the crisis, headline inflation is just above zero in both countries today. Core inflation rates are basically identical, very tight labour markets around equilibrium. Lack of job creation shows unemployment at around 5%, steady improvements in the labour markets turn the slower headline wage growth inconsistent with these other measures. Wages have, if you adjust them for hourly inaudible real wage growth adjustment for inflation, wage growth isn’t quite as weak but there is still puzzle there. So that’s where the Labour markets are very similar on all of those measures.
Now before I move on to the differences in the two labour markets, this is the time I am going to quiz you, see how many of you know the data very well. Dover, Exeter you can have a guess, who thinks Dover is the UK? Who thinks Exeter is the UK? If you were my students I would make you all vote. I’m not sure I can read much into that. I am now going to take you out of your Dickens like misery and show you which is which.
I am going to show you labour market participation in the two countries, in the UK and the US or Dover versus Exeter. Dover shows a striking pattern where labour market participation, how many people are engaged in the workforce either working or looking for work. You will see labour market participation is basically flat since 2007. So basically about the same share of people are working in the population as 2007. And the other thing that you are going to see in Exeter is a sharp fall in labour force participation because there is a significant fall in the number of people in the country who are actively working or looking for work. So how many people think that Dover is the UK? So Exeter is the US and this is the graph that would make that very clear to those of you who have at least followed this debate. In the UK this black line here is labour force participation. It’s quite striking basically the same share of the population working now as working in 2007. In the US it has dropped by about 3.5 percentage points and that’s a big fall in the amount of people that are working or looking for work. So what explains that? Some of what’s happened though is a natural demographic affect in both countries and that’s in the dark blue and light blue bars. That just captures the fact that the UK and US are ageing. As people age they do tend to work less as people get closer to retirement age or around retirement age, older people of retirement age do work less. So some of that is just a natural decline in participation and that’s happening again in the blue bars in the UK and the US. It’s a little bit bigger in the US than the UK, about 1.8 percent decline in participation just from natural ageing and about 1.5 percent in the UK. So that’s roughly the same. And also we have a trend in the both the UK and the US which is happening around the world which is more older people are actually working longer and staying engaged in the workplace for longer. This is probably because people are healthier they can work longer and maybe the changing nature of work. We are seeing a bit more of that in the UK, older people are staying in the labour force a bit longer. But that’s not the main difference. The big difference is the yellow bars of the two countries – the infamous other that always shows up in these graphs. And what explains the yellow bars? Why is it that a big chunk of the middle age and when I say middle age I mean the 16-65 year olds, working age population, in the US a proportion of that chair are just not working they are dropping out of the workforce. In the UK you have seen an increase in those aged 16-64 in the workforce. That’s the major difference. There are a lot of factors behind that, I will talk about some of those later today but one factor is probably changes to the benefits policies which have made it more attractive to stay engaged in the labour force in the UK for certain workers.
The bottom line is one of the most striking differences between the UK and the US labour markets is this change in participation. In the UK participation has been pretty flat, in the US there has been a sharp fall in the share of the population participating in the labour force. The second big difference between the UK and the US is self-employment. The top of the line shows the share of the working force population that are self-employed in the UK, the blue line on the bottom shows the share for the US and a couple of things pop up for me. First is that the share of self-employment in the US has been quite a bit lower than in the UK historically, maybe I’m biased to my team who are always calling companies but in my head I had this image of the US entrepreneur in groups of men who would start a company out of garage whether it is Microsoft or Amazon or Google or Harley Davidson it grows into a successful company and this is in the head of the Americans. If you have some free time, go and google companies started in a garage. You will first of all be shocked by how many websites there are which will tell you how many companies were started in a garage but you will secondly be surprised by how many of those companies are American companies, so it reinforces the sense that in America there is more self-employment and entrepreneurship, but that is contradicted by this data. The fact that in UK you have higher self-employment historically and you’ve had a bigger increase, and this is what’s really striking, quite a sharp increase in the past 5 years in the number of people in the population who are choosing to be self-employed. Not all of them people who are self-employed in the UK are starting the next Google or hopefully not working out of a garage, a lot of them are people who just run small companies many in the construction industry, many are professional industries were they contract out their skills on a core contract basis for consulting or drafting of professional services. But nonetheless this is a pretty striking trend in the difference in the labour markets in the UK and the US, more self-employment and a bigger increase in self-employment in the UK.
The third difference between the two labour markets is in migration. This graph shows the net migration is a share of the working population in the UK and the US. The UK is in red the US is in blue. We see that the migration share is higher in the US for much of the 1990s and early 2000s and then starting around 2005 the migration share in the UK has been higher than in the US. The big differential between the two has fostered around quite a bit, basically identical migration shares as of 2013 but quite recently the UK migration share has increased quite a bit, more sharply so that now net migration to the working age population is expected to be about 0.4 percentage points higher in the UK than the US IN 2015.
One final difference between the labour markets in the two countries and this one is a bit more speculative, it’s not quite as clear in cold data, but I think it’s one that is worth thinking about and might have some interesting interactions with some of the other differences between the US and the UK, and that’s the consistency of the recovery in these two labour markets. So this graph, some of the charts which I couldn’t show you earlier are some of the measures of labour market churn. In the UK there and in the US right there and this again shows how comfortable workers are moving from job to job, finding a new job and how many vacancies companies have for hiring. If you look at the graph from the US for these four different measures which altogether captures efficiency in the labour market, you see how it declined in the crisis but then you see things turn around 2009/2010, and you see pretty straight upward lines in each of those graphs. I am not going to do fancy metrics but you can pretty much draw a trend line for each of them starting in 2009. That shows a pretty consistent recovery year after year in each of these areas in the US. But they started to do the same thing for the UK. They started to turn around 2009/2010 and then there’s no steady upward lines there. They go up, they go down, especially around 2010/2011. Around this bright red line is quits, initially workers were comfortable with quitting their jobs and getting other jobs, and then they got more nervous again and stopped quitting and quits fell back down, showing increased concern about the stability of the recovery in the UK.
One other graph which suggests recovery in the UK has had many more stops and starts. This is called a leverage curve. This is a nice graph which captures how efficient a labour market is functioning. It looks at the unemployment rate on the bottom, you can see right on the vertical axis it captures how many workers are unemployed tightness from the employees at the bottom and then it captures tightness from the people who own the companies at the top, how many openings they have got which they are trying to fill and hire. Traditionally what countries do is during recovery they usually upwards so at the top left, there’s not many low unemployment rate and a lot of vacancies in companies. As an economy slips into a recession it goes down the curve because unemployment goes up and there aren’t as many vacancies to be filled and then when it recovers, it sort of starts to move back up. So you get unemployment slowly falling so you move that way, we have got more and more vacancies when are companies are hiring so it goes that way. The pattern of the US is pretty typical during a recession you move down, during a recovery you start to move up and to the left. You see how steady it is, every colour is a different year and each year is based on a similar process in the recovery is more or less a straight line. Then let’s take a look at what happened in the UK – seen a fall in the recession and then the recovery started to turn around and then it got stuck here 2010-2012 it just sort of bounced around here, in the US it was steadily going up. Finally the recovery came to the UK in 2013 and has since followed the normal pattern. I could show you number graphs like this but it makes the same point that in the UK labour market recovery has been very strong the past few years, basically identical to the US, but it took a bit longer to get going there were some stops and starts in the recovery in the UK starting from about 2010 going up to about 2012. So it does make you wonder if there are more people in the UK who are more worried that the recovery hasn’t been as grained for long, it’s less consistent, so less faith in the recovery than you would in the US where it’s been going on longer.
So four major differences between the UK an US labour markets in terms of more participation in the UK, more self-employment in the UK, more migration in the UK but a little less consistent stable recovery for not as long in the UK. There are probably some important linkages in these four differences in the UK and the US labour markets, these are not all independent factors they are related. In the written comments go into a lot more detail about this, I will run through them quickly and spend a little more time on the one which I think is the most interesting. One important linkage is probably what has been going on in Europe with the Euro crisis. We obviously see that more tightly linked to the UK than the US, the Euro area has obviously hit some challenges for the past few years especially around 2010-2012. Those challenges probably led to those sort of stops and starts in the early stages of the recovery in the UK but didn’t affect the US as much. Also the challenges in the Euro zone – slower growth, slower employment growth which probably helps in the increase in migration to the UK. Another link, in the UK there is higher self-employment and higher participation and those are probably linked. If you break down who is self-employed and the increase in self-employment that’s come in the UK, you will find that a decent share of that is from women and older workers aged 65 and up. Some of those workers probably don’t want fulltime, intense 9-5 Monday to Friday so the fact they can become self-employed lets them enter the labour force and participate in the workforce. So this trend, more women, more older workers, workers who are aged over 65 being able to become self-employed and thereby participate, is probably driving both of those trends in the UK.
The final factor which is a point in some of these trends is migration. The increase in migration in the UK recently is probably also linked to increased self-employment and increased participation. I actually have a few graphs which I think are particularly interesting on that, let’s start with self-employment. This graph shows the likelihood of being self-employed in the UK by your citizenship. Blue is if you’re a UK citizen, green is if you’re not a UK citizen. We see that a big reason for the big pick up in the tendency to be self-employed is in migrants. Migrants have shown an increased tendency to be self-employed in the UK. So if you just do some simple back of the envelope calculations in basic algebra, there’s more migrants coming in to the UK, and migrants are seen to have an increased chance of being self-employed we can calculate how much of that is feeding into the data for self-employment in the UK. When you do the basic algebra this is what you find. The black shows the increased in self-employment in the UK, one of the big differences with the US labour market. If you break it down into how much that increase in self-employment is driven just by having more migrants that’s there in green it’s pretty tiny, but then the red shows the fact that migrants are becoming more likely to become self-employed. And that’s driving, not all for sure, but a decent, measurable, substantial part of the increase in self-employment. So increased migrants again are one factor driving the increase of self-employment in the UK.
Now where the role of migrants is even more striking is when you look participation if you compare the data for the UK and the US which makes it stand out even more. The set on the right is the participation rates in the US, in green is participation rates for immigrants, the blue is participation rates for Americans, native born Americans. We see both are falling. Immigrants are more likely to participate in the labour force, but again both are going down, which is what you see for the US as a whole. Participation has fallen sharply in the US as a whole which is again one of the big differences with the UK. In the UK, for native UK citizens, participation is pretty flat which is what we see for the population as a whole, but participation rates for immigrants have been going up by quite a bit. There much higher and increasing. Migrants in the UK are more likely to participate in the labour market and that rate has been increasing. So now if we do that basic algebra again, more migrants into the UK, more likely to participate as a base case and increasing participation overtime, over the last few years, how much of that drive has increased participation in the UK vs the US? And this is where it is really striking. Again in the US, big black line big decline in participation. That line is largely from its native citizens, Americans, in blue a little bit of the decline is also from migrants who have been less likely to participate. But then where it is very different is in the UK. In the UK, UK citizens in the dark blue have been less likely to participate overtime. That’s largely a demographic affect, the UK population is ageing, but the red and green bars show the increase in the participation rates in the UK, and those are directly related to migrants. The green is simply the fact there is more migrants into the UK and migrants have a higher tendency to participate in the labour force. The red shows in increase in tendency to participate by migrants, and if you actually put those two together, that really balances off some of the demographic effects and other effects in the UK. It is critically important in driving the increase in migrant participation in the UK labour market than the US.
I could keep going with lots of details like that, but I think it is time to take step back and see the big picture. What does all of this mean for wages and inflation and monetary policy in the UK? So let me now try to tie this together. We have seen striking similarities between the UK and US labour markets but also some key differences in participation, self-employment, migration and the durability and length of the recovery. So I just want to take a step back and say why are we looking at all those details on the labour markets, why do all of these factors matter so much? The main point of all this is to think where monetary policy should go and where inflation is. Labour market developments are critical for predicting where inflation is going in the future. Inflation is very hard to predict, it’s very volatile. Inflation is also very much affected by external shocks, today this is very probable. 7/8ths of the reason why inflation is at 0.2% instead of our 2% target is because of sharp falls in energy prices, food prices and imported goods prices. So if you want to know where inflation is going in the future, you need to try and get at what is the consistent underlying trend rate of inflation in an economy. A key determinant of the underlying inflation rate in an economy is wage pressures and new labour costs and labour worker pressures. So that’s why we are going through this exercise. If you want to understand where inflation is going, you need to understand this more permanent, consistent trend of inflation which hinges critically on the labour market. It’s also important to do this because when you set monetary policy and adjust interest rates, it can take 1-2 years before any change in monetary policy heeds through and has its full effect on the inflation. So that means we can’t wait for inflation to take off today and then adjust monetary policy or we will be too late. We need to try and predict where inflation is going in the next two years and adjust monetary policy based on where we think it is going. Traditionally the best way of doing that is again back to understanding the labour market and the pressures in the labour market.
So that’s the goal here. So to try and sum up, I have just shown you a ton of data, ton of charts and a ton of graphs. To try and sum it all up, what I did behind there was I took each of the different statistics I showed you and gone through, including some in the written version which I didn’t even bore you with today, since we have seen enough and then I colour coded each of them. I colour coded them red if that indicator suggests the labour market is already at full employment, at the risk of overheating, at the point where wages usually take off. Orange was colour coded as a labour market is in a trend, it should be there quite soon. Green was errors were there might still be a little bit of slack left in the labour market, suggesting wages were about to take off. If you did that with all these indicators, basically the whole chart was red, suggesting tight labour markets, wages are going to accelerate, not much more room. The only small errors you might have identified some slack, where in the US labour market, unemployment may have a bit more to fall and there may be some space in participation, so many people have just stopped working, dropped out of the labour force in the US, if they came back in they might be a bit more slack in the US. There is a big debate over whether that will happen or not. Then in the UK there might still be a bit of slack just because the recovery hasn’t been going on as long, it’s not as well engrained, there’s been more fits and starts, people may not be quite as confident in the recovery. But that’s really pushing it to find any evidence of slack in these two labour markets. By economic measures these labour markets are red hot, this is where wages start to take off. But as I also alluded to before quickly, the catch is wages aren’t taking off. There does not seem to be a lot of movements with wages. So there is a lot of different measures of wages to look at. Let me just summarise. The two most popular measures of wages are looking at average weekly earnings and if you look at those measures every inaudible in the UK, most recently have been about 2% or 2.2% depending at what measure you look at, about half of where they were before the crisis. In the US also, average weekly earnings are well below where they were before the crisis. And what is particularly puzzling is even though is they are still below crisis levels, they should have started to increase, they should have started to get momentum and that hasn’t happened in the UK. As at the end of the third quarter which goes more 3/3.4% and if you look at the last 3 months through November, the last time we had data on them, it’s down to 2/2.2%. Wage momentum seems to be stalling rather than picking up. Even heading in the wrong direction, not only is the level low but we are heading in the wrong direction. So why is this? This is a puzzle which we have spent a lot of time thinking about it and it is really critical to understanding what is going on in the world of the labour market. So there’s five main reasons as to why we think wage gains may be slower than one would expect.
First could be the changing composition of the workforce. More people are getting jobs now who are younger, less experienced so of course they earn less and that can slow down wage growth. We have some estimates, at the Bank of England were we breakdown these effects and we estimate that that could of explained 0.8% of the percentage point in the weakness in wages in the third quarter. This is a real factor slowing down wage growth in that that is also likely to be temporary and not consistent. There also could be the fact some people are working less hours and that’s true in both the US and the UK, some people are paid on an hourly rate so it’s not surprising that if you work less hours, your total wages are going to be lower. Third factor is since headline inflation is low, that means that companies are paid as vast wage gains because really things are going up, things are cheaper, so that may be way down in inflation. That would also fade as inflation starts to pick up. It could be that there is some slack in the labour market which we are missing or we are not measuring it the right way and unemployment could fall quite a bit more and that might be why wages are not picking up. Finally there could just be changes in the structure of the global economy which has reduced the bargaining power of employees for example less trade unions, more globalisation, factors like that. I’m sure more than one of these factors explains weak wage growth but there is some evidence that at least the first three have played some role and you can see that just from looking at other wage measures. Let me show you, there are a few more wage measures. If you look at earnings adjusted for hours worked in the UK, audience growth will increase from that standard measure of 2% annual wage growth as from that last 3 months of data, due just to the fact people are working less hours, wage growth increases to 2.6%. So that makes up some of the gain, not all of it but it makes up some of it. If you then adjust for the change in composition for the workforce, that workers are younger, less skilled, less skilled jobs, that can actually increase, we don’t have data yet but as of most recently in the third quarter, that can add another 0.8% on wage growth. So we can again close some of the gap of what you would be expecting to see. If you were adjusting for low inflation today, look at say the real consumption wage, just below inflation, we could get inflation increasing from about 2/2.2% to 2.9%. So then we are on a little bit more. So when we look at all of these different ways of measuring wages, they do matter. We do need to think very carefully about what measures we are looking at. We look at some of these different measures, wages aren’t as dismal as some of the headline numbers may suggest. Though some of it although it is still stronger, bust is still below what you where you would expect given how tight the labour market it. That’s captured if you average all these things together. If you take all these things into consideration and average wage growth you get is still about 2.6% versus a pre-crisis average of 4%. So there is still some puzzle there, wage growth is still not as strong as you would expect in this stage of the recovery. The US is facing the same problem as over here, average wage growth in the US, even after adjusting for all these measures, still closer to 2.5% versus the pre-crisis average of 3%. And by some measures, wage growth in the US has actually been lower than in the UK. So what does this mean for monetary policy? One more set of data and then I will tie it all up.
If you want to think about how all of this matters for overall inflation and monetary policy, you need to look past just wages and adjust for productivity. If workers are more productive, you can pay them more but it won’t pass through into inflation. So you really need to adjust wages and adjust earnings for productivity. Last chart I promise. It’s a different way of measuring wage costs and unit labour costs to adjust for productivity. Don’t worry about all the details, the bottom line averages, I will give you the key point, is what it suggest is that if you adjust for productivity and look at how much average production costs, wage costs are increasing for companies, they are going up more than the headline wage numbers suggest. For example unit labour costs in the UK at the end of the first quarter were only 0.8% compared to the third quarter they were 1.9%. So they more than doubled. The US were also going up, 1.4% at the end of the first quarter, and 1.5% at the end of the third quarter. So unit labour costs, the cost of producing based on labour markets, is going up and it’s accelerating. Although it is still not where you would expect given the tightness of the labour markets. So still below pre crisis averages. The reason it’s going up is not because wages are going up more quickly, it’s largely because productivity is very weak, especially in the US. In the US unit labour costs, average growth of 2.5% which is actually above pre crisis averages. But the main reason is not student wage growth, which we saw growth was even lower than in the US than the UK recently, it’s the fact that productivity has been so abysmal in the US in the last quarter. So what does all that mean? Where we tie it all up. First just let me say congratulations, I’ve just whipped you through a ton of charts, a ton of data you might just feel like you have read a long Dickens novel, hopefully you weren’t inclined to snooze, but it has been a lot of data which I have took you through but I have done it because looking at this data is really critically important to understanding what is going on and what the puzzles are, that it is not straight forward. So I want to summarise on 2 points – what does it mean for the UK vs US in monetary policy and what are the implications just for the UK.
So first, UK vs US monetary policy. I spent a lot of time on the similarities between the US and the UK labour markets, striking similar and it may be tempting to say based on that the US and the UK should be doing the same for monetary policy. I won’t say which one is right and which one is wrong but let’s just say they both can’t be right if you focus on the similarities in the labour markets, so I don’t want you to draw that conclusion, when you set monetary policy you need to look at something more than just the labour markets. And there are some other pretty fundamental differences between the UK and the US, for example the US has a much looser fiscal policy which is going to act as a spur to growth in inflation in the US, the UK has a tighter fiscal policy which is going to act head with to inflation and growth. Also the US does not have macro prudential policy at a central bank such as the financial policy committee The Bank of England, so the US may have more incentive to raise interest rates than just overheating in specific sectors such as the housing market, while in the UK we have specific tools of macro prudential policy which can address that. On the plus side, the US has faced a sharper, larger, currency depreciation recently which will weigh down growth in the immediate future more than in the UK but that might be a reason why the US might grow slower. But you put all those factors together and what it highlights is the labour market it key for monetary policy but you also need to look at some of the other aspects of the economy. So I don’t want you to leave here saying the two countries should follow the exact same path of monetary policy as there are some clear and fundamental differences in their two economies. So what does that mean for monetary policy in the UK? This is my attempt to sum up all what we have talked through today. Basically we started off by talking about labour market economies, unemployment rates in the UK and the US suggested very tight labour markets. Traditionally what we do is start in the inaudible labour markets by looking at that, because if you have tight labour markets, as we have seen, then that will feed through into higher wages, higher unit labour costs which in turn feed through into higher inflation. There are very tight labour market economies which would suggest stronger wages, stronger unit labour costs but that chain has somewhat broken down. Tight labour markets are not feeding through in to the stronger wages the way you would traditionally expect. I did show you some other measures of wages which would suggest it’s larger than other measures so some momentum but still not the amount of momentum you would expect at this point in the recovery and given the tight labour market costs. Also I showed you there has been some breakdown in the slip between wages and unit labour costs. Traditionally wages went up a certain amount, you knew how much unit labour costs would go up because productivity was pretty stable, but productivity has broken down, it has been pretty volatile and really hard to understand and predict. So this relationship between wages and unit labour cost has in turn broken down and there’s also been some research recently showing the relationship between wages and unit cost of inflation is weaker than it has been historically. So these three costs are all broken down and their connection is not as strong as in the past. So some people would suggest that we just ignore the middle cost and go back to basics and just focus on what’s happening with labour market economies and that will be a more consistent predictor of where inflation will be over the medium term, consistent underline inflationary pressures. There’s some evidence that even more importantly looking at things like turn verbal’s may even be more important that the labour market getting tighter since these relationships are breaking down. But this just highlights the puzzles which are out there. But the key thing is, in the tale of two cities, the key theme is resurrection, coming back to life. And there is no doubt that the labour market in the UK has come back to life. It is a labour market which is largely normal based on the labour market quantities but these are for some reason not passing through into wages and unit labour costs in the way one would expect. So what does that mean for monetary policy? The last point, the big question of the day. So for me, what it means is that tightening monetary policy would require faith that our forecasting models based on this set of assumptions, work, and that the tightness in these labour market economies, in measures of labour market turn will soon translate into higher wages and unit labour costs, which in turn will translate into higher inflation. My personal view is I still have faith that these relationships will hold but I do want to see a bit more upward movement in these waging costs measures in the middle to build confidence that this chain of tight labour markets feeding into higher wages is still intact. In other words, I believe you should trust the models they are verified. But on a brighter note, the recent falls in oil prices by delaying the recovery in inflation do provide the luxury of a bit more time to evaluate if this chain of linkages is working and to build confidence that this chain is working. Once this upward momentum in wages and unit labour costs builds, which I expect it will, it will be time for the UK to follow the example of Ireland and begin the slow and gradual process of tightening interest rates. So on a brighter note, the relatively smooth experience of UK lift off in raising rates, will suggest that unlike in the tale of two cities, we will not have a revolution when that day does occur. And with that I will stop and I will be delighted to take any questions.
Well thank you very much and because we haven’t got much time, can I just ask a really quick question. So when do you think interest rates are going to go up?
When wage and unit costs start to get a lot better, if you can tell me when that is I will have an answer.
You were talking about participation rates in the US but what I don’t understand is its part of the political inaudible and that’s what makes it so interesting and I feel it needs to explained a little bit more because I still don’t understand still, in this country where we have had welfare for unemployment but in the United States it doesn’t quite work like that, obviously there is that aspect and I was wondering if you explain those different statistics suggesting that an average earnings and family earnings are still looking several thousands of dollars below the line inaudible pre crisis whereas in this country they have actually gone inaudible to what extent is that an impact?
Inaudible I just wanted to hear about any changes in the US economy inaudible I am just wondering if you think that inaudible.
I just wanted to hear your thoughts on inaudible …
Can I sort of exercise more on this, what he said is that he the recovery of household incomes since 2008 there was a real difference in age groups, the over 65s seemed to have recovered and the under 35s are really struggling. What sort of category of shifts does that fall into?
Some great questions, so let me see how many I can get through. So first participation rates in the US and the UK. So you’re right this is a fascinating debate in the US, there are so many heated discussions which encompasses on is this drop in participation permanent or temporary. Some work by Alan Kruger who one is the most respected labour market economists in the US, suggests that a lot of the people who dropped out of the labour force in the US are not coming back. He shows that a lot of them have been out so long that they are discouraged, they have lost skills, and they’re not coming back. There’s also another line of literature which argues that there has been some substantial changes to the benefits policy in the US which provides less incentive to get back in to the labour market and a particular one which has gained a lot of attention is disability. In the US it is pretty easy to apply for disability. Disability insurance in the US, people taking up disability has been a shocking upward trend that you don’t see in the UK. The UK has tightened up on eligibility, in the US if you get disability, you qualify once, you can collect it for life so there is very little incentive to go back in. In the UK they check in and you need to keep qualifying, there’s more checks. So there are a number of changes like that what add up to make more of a difference. One just more fun antidote on this is, when I was back in the US one day we were off from school one week and there was a bunch of twentysomethings skiing, going up the ski mountain and of course I ride up the lift and say what do you do and ask them about their employment, but most of them were on disability insurance and they were all skiing! And they were all in some clubs were you would go online and it would tell you what to fill out to collect it, and then you could collect it, ski, and a lot of them blogged at night under the table. So those types of things are just not cut up in the US in terms in the increase of participation. Some people do want to get back though and they can’t. And also while it is easy to be cynical, a lot of people have written off these people coming back into the labour force but the last quarter of data released in the US shows that some of these people, the one’s we thought had dropped out are coming back to the labour force, so there probably is a little bit more space there, what we are probably going to see if something in between those two extremes.
Pre crisis earnings, yes there are different ways to measure it, whether it is exactly above or below what is what before the crisis depends on which measure you use. But the bottom line is earnings haven’t gone up for a decade in the way you would expect, especially considering that the economy has been engrained for a number of years, so there is still some catch up there.
Productivity growth, critically important question in all of this. Obviously if productivity growth picks up and wages pick up, it’s still a long time before you need to raise interest rates. In our forecast there was a pickup in productivity growth over the year, and we got excited it was finally here, but it looks like fourth quarter productivity growth is going to be abysmal. That’s part of the trade-off of strong employment growth, with moderate overall GDP growth which generally turns into weaker productivity growth, there is some trade off there. So I think the fourth quarter figures for productivity growth are not going to be pretty, but then again productivity bounces up and down every quarter and I don’t think any of us read anything into one quarter, we need to look over the trends to get a sense of what underlying productivity is. If it does not pick up as we are expecting, then it will mean that even moderate wage pressures will feed into inflation faster. So we are all going to be watching the trend, not just one quarter, but the trend very closely.
In terms of seeing if the current inflation is weighing on wages your exactly right we also do look at inflation expectations and if that is weighing on wages and there’s some ways of doing the analysis you do find that inflation expectation rates are more important than current inflation. But having estimated many of these relationships over the last couple of months, you will see that I didn’t include any regressions in my comments and that is because they are so highly unstable. My staff sitting here can tell you, who have run off more of these than they have probably ever wanted to, you can sort of get whatever you want depending on how you run it when you change the specification. The relationships really broken down and that is what makes this so much more difficult to understand.
Can statistical agencies keep up with measuring self-employment? This is another big debate especially within the US are we measure technology, gamers, probably not but I haven’t seen anyone else suggesting that we are measuring self-employment for those reasons. I think some of the big issues are, are we mis-measuring productivity growth, this is for me the biggest question.
That super cycle, not quite sure what you was getting in to there but my take on it is that levels did get too high before the crisis and they have to come back down. In the US, debt levels are quite manageable in moderating, there still a bit higher especially in personal levels than in the UK. But we are getting to a point where it is more normalised.
Finally recovery in household incomes. Different age groups have had very different effects through the crisis but what I found interesting is that some of the effects in the labour markets are the opposite of what you are talking about. There has been some more recovery in wealth, in that older part of the generation, so say 55 and up, 65 and up have seen more of a recovery in their earnings and their wealth. Yet they are some of the people who have seen the sharpest decreases in the UK labour force. So we used to wonder if it was because especially with people near to retirement it showed a big hit to their earnings, that’s why they were just going to work a couple of extra years to make up for that loss in earnings before they retire but were not seeing that, were still seeing much more engagement by people in the upper end of the demographic despite the fact they have done ok in terms of making up a larger part of the losses in their wealth. So some of the trends work in different ways but there is this breakdown by age group is very interesting.
Thank you very much indeed and we are now clearer as to why economies are not clear (laughter).