Examination of Witnesses (Questions 40
TUESDAY 9 OCTOBER 2007
Professor Stephen Nickell
Q40 Lord Sheldon:
But when they come in they get jobs. There must be some calculation
as to the kind of job that they get.
Professor Nickell: Yes, there is. Migrants overall
get jobs at all levels. Basically, the skill levels of migrants
are remarkably similar on average to the skill levels of the native
population. I think slightly more migrants have degrees than the
native population, but by and large the education levels match
up quite well. However, because over time there is a persistent
move in demand away from unskilled towards skilled, it is always
in these circumstances that the unskilled part of the population
tends to get worse off, and insofar as there are ethnic groups
who tend to have lower skills than the average, they would be
most affected by this.
Q41 Lord Layard:
How do you think the existence of a national minimum wage administered
by law affects the employers' demand for migrant labour compared
with domestic labour? Do you think that the existing policies
ensure employer compliance with minimum wages for migrants?
Professor Nickell: That is quite an interesting
question. I am not sure I am the best person in the room to answer
it. First of all, insofar as unskilled migrant workers are willing
to travel to the UK and work for wages below the national minimum
wage in the UK, clearly the existence of the national minimum
wage, insofar as it is enforced, would reduce the demand for workers
from that group. On the other hand it is worth bearing in mind
that unskilled workers, ie, those who have no qualifications,
are a relatively small proportion of all migrants, so I am not
sure how significant this is. I fear I am not really an expert
on compliance and enforcement with regard to the minimum wage.
We do know, of course, that there are numerous stories in the
press and elsewhere about migrant workers in agriculture and construction
who are not well treated, but most migrant workers do not work
in these areas. I guess my overwhelming answer to this is that
I am not familiar enough with the compliance and enforcement literature
on minimum wages to be able to answer this question very satisfactorily.
Q42 Lord Turner of Ecchinswell:
Can I return to the question as to whether the empirical evidence
is in line with what we would theoretically expect, given that
the theory would be that if you get a flow of new factors of production
not matched by others, ie, more labour not matched by more capital
or more unskilled labour not matched by proportional quantities
of skilled labour, you would expect to see perhaps a short-term
employment quantity effect but a long term relative price effect?
You seem to suggest that we are seeing that because we are seeing
some price effect, but is it your opinion that those price effects
are of the quantity that one might reasonably expect, given the
theory, or are they strangely smaller than one might have expected,
given the theory? Is there any disconnect of empirical and theoretical
approaches here or is it to your mind a fairly consistent story?
Professor Nickell: In the literature most people
are relatively surprised that the wage effects are not as big
as they expect. A lot of this literature started in the United
States, and a good early example was an analysis of the large
influx of Cuban migrants into Miami quite a long time ago. When
the wage levels of workers who would be expected to compete with
these migrants were investigated before and after the movement
of migrants the changes were very minimal. Some argued that one
of the reasons for this was basically that if you analyse these
things by region, by area, you do not take account of the consequent
effects on the regional migration of native workers, so people
argued that what happened in Miami was that a lot of native workers
either moved out of Miami or did not go to Miami when they would
otherwise have done so. George Borjas, who is a famous Harvard
economist, espoused this theory and, using analyses which take
account of this, he does find bigger wage effects. However, what
I would say is that anyone who has tried to replicate this analysis
in many other countries has never found big wage effects. This
is an ongoing controversy but I think still, by and large, people
think there are small wage effects and, furthermore, they do not
think the small wage effects they find are due to responsive native
migration around the country away from the immigrants. What is
the answer? The best answer that people can come up is that in
areas where, for example, in-migration has generated a large excess
of unskilled workers the technology and the capital adjust very
rapidly to take account of this. Putting it in the crudest possible
way, lots of sweatshops suddenly appear and in some sense the
capital and the other factors seem to adjust rather rapidly.
Q43 Lord Turner of Ecchinswell:
So the capital moves from somewhere else?
Professor Nickell: Yes, and it is also of a
different kind. It is of a more labour intensive kind if you have
lots of unskilled workers. It has to be said that this is ongoing
stuff and no-one has quite got to the bottom of it yet.
Q44 Lord Griffiths of Fforestfach:
Is there any empirical evidence to suggest that as we have a flow
of immigration so employers in the private sector and the public
sector invest less in training and in labour-saving devices in
Professor Nickell: As far as the evidence is
concerned, there is some evidence along the lines which I just
mentioned, namely, when there is a big influx of unskilled workers
into a particular area there does seem to be more response in
the type of capital that is employed and that capital is fundamentally
at a lower technical level than in areas where there is a ready
supply of skilled workers than of unskilled workers, but most
of that evidence is from the United States. The thing about the
UK is that the skill mix of migrants is so similar to the skill
mix of the native population that the training incentives are
not affected that much, at least in the longer term. Of course,
if there is a surge of migrants, some proportion of whom are skilled,
then the employers may be tempted to go out into the market rather
than do their own training, although we know there are a lot of
other issues at the moment about why they might go out into the
market instead of doing their own training anyway, but ,leaving
that aside, there may be these short-run effects. My reading of
such evidence as there is and my understanding of what goes on
is that because of the similarity between the pattern of skills
in the migrant workforce and the native workforce there is not
much of an impact over the longer term on training incentives
and incentives to invest in particular types of capital.
Q45 Lord Skidelsky:
You have partly answered this, I think, in an earlier reply, but
I want to ask you what effect you think immigration has on the
rate of inflation. In your experience has immigration been a significant
consideration for the Monetary Policy Committee?
Professor Nickell: The answer to the last bit
of your question is yes. The first and obvious effect with migrants
moving is that there is an inflow. Probably that temporarily increases
the degree of labour market slack, puts down the pressure on inflation
and the monetary policy response. That in some sense is not very
troubling because we can respond to the inflationary consequences.
Where we found in-migration troubling was that what we wanted
to do was to try and get good estimates of the potential output
in the economy going forward, and in particular the growth of
potential output, and the growth of potential output is basically
productivity growth plus the growth in the number of people, and
it was the latter where we had trouble because the estimates of
projected in-migration, which was a key part of the projections
of the growth of the labour force going forward, first of all
kept changing all the time and, secondly, they seemed to be based
on a method of collection which did not appear, at least then,
to be wholly reliable. Since the growth in the potential output
is a key part of monetary policy making because we are interested
in that relative to the growth in demand in the economy, and if
you do not have a very clear idea of what the rate of growth of
potential output in the economy actually is, your forecasts of
inflation become that much more uncertain. We used to worry about
this quite a lot and we had people at the Bank going to talk to
people at the ONS about how the data were collected and so on,
and this was quite a serious issue for us.
Q46 Lord Skidelsky:
The task would have been easier had the numbers been better?
Professor Nickell: Yes. I do not want to go
too far in this and say it was really a disaster because we just
did not know what was happening, but yes, it was somewhat frustrating
in the sense that we spent a lot of time looking forward and,
of course, you do not know what is going to happen in the future,
but not knowing where you are at any given time is rather frustrating
because it does not just happen for migration; it happens for
GDP and all sorts of other things as well.
Q47 Lord Griffiths of Fforestfach:
From your experience of the Committee do you think there was ever
a time, as you look back over the decisions you made in setting
interest rates, when, if you had had more correct forward data
or better predictions going forward, you would have set a different
Professor Nickell: That is a good question.
Q48 Lord Griffiths of Fforestfach:
Because that, it seems to me, is the real rub. I see intellectually
you make a case, and if I were on that Committee I would feel
exactly the same as you did, but the rub is, did you set interest
rates on the numbers?
Professor Nickell: Let us put it this way. We
may have made mistakes but I do not think we made mistakes because
Q49 Lord Kingsdown:
I have another question on the impact of immigration, this time
on housing and on house prices, both in the private housing market
and on social housing.
Professor Nickell: Since the late 1990s, as
far as we know, net migration to the UK has been around 180,000
a year, which is considerably faster than it was in the past,
and that has probably contributed to between a third and a quarter
of the rise in the number of households over that period. House
prices over that period, as we know, rose very rapidly, and the
drivers of the rise in house prices have been the continuing rise
in household incomes, the rise in the number of households, and
the comparatively low levels of house building. I think, of these,
the rise in household incomes is the key. We have quite good information
on the impact of changes in the rate of increase of the number
of households on house prices and looking at historical data we
know something about that. So, to give you an example, doing a
bit of a back-of-the-envelope calculation, I found the following:
if there had been no net migration since 2000, then instead of
house prices rising over the period from four times average earnings
to seven times average earnings, the calculation I did suggests
that house prices would have risen to around six and a half times
average earnings. So that is not insignificant but it is not in
any sense the story. So that gives you some idea of the market
housing sector. What about social housing? In the big picture,
the demand for social housing is actually driven in quite large
part by the price of housing and the level of rents in the market
sector, and as these have gone up so the demand for social housing
has risen. That is very important. What particularly about migration
directly, though? We know that some migrant groups, such as Somalis
and Bangladeshis, are much more likely to live in social housing
than the average UK individual but, perhaps more importantly,
the vast majority of migrants are far less likely to live in social
housing than average. Of course Indians, who are currently the
largest migrant group, are only half as likely to live in social
housing as the average UK citizen and the same applies to Poles.
Of course there will be some groups of migrants who, for a variety
of reasons, make particular demands upon social housing; but,
overall, migrants are less likely to use social housing than the
Q50 Lord Turner of Ecchinswell:
My supplementary was whether you had looked at what the balance
between household fragmentation and total population growth might
be going forward. You said that over the last 15 years or soI
think it wasthat a third of the increase in households
was coming from net migration.
Professor Nickell: Yes, that is what the ONS
Q51 Lord Turner of Ecchinswell:
And about two-thirds from household fragmentation, ie smaller
average size of household?
Professor Nickell: Of existing households.
Q52 Lord Turner of Ecchinswell:
Presumably, however, the household fragmentation process, that
people live in smaller and smaller households, is an effect which
at some stage comes to an end, so beyond a certain date the proportion
which is actually driven by population expansion would become
bigger, would it not?
Professor Nickell: I would think it is going
to go on but one of the key features of this is the fact that
people live for longer periods. Life expectancy is rising linearly
and has been doing so since 1870. Despite projections at every
point that it was going to level off, it never does, and of course
if people live longer, and longer and longer, that of itself raises
the number of households without anything else having to happen.
Q53 Lord Turner of Ecchinswell:
Because it drives population growth?
Professor Nickell: Exactly, but also of course
since old people rarely live in more than two-person households,
and often in one-person households, that is a very important factor
which is probably not going to level off. Certainly in the projections,
there is no levelling off in this process of households getting
smaller and smaller. Of course, we know that projections in this
world are often extrapolations of existing trends and therefore
you could say, well, of course, because that is the way they project.
There is no real sign of it levelling off but I take your point.
Q54 Lord Skidelsky:
Maybe I could smuggle in two very quick questions. Your previous
reply on the demand for housing suggested that perhaps a disproportionate
demand from Indian migrants was for private rented accommodation.
Professor Nickell: No, I said they did not use
social housing. In fact, I looked at the numbers and 86% of Indians
live in owner-occupied houses, by far the highest group of owner-occupiers
in the whole country.
Q55 Lord Skidelsky:
That is very interesting and the second question, which no-one
has asked but is on the menu, is: do you think the Government
should curtail immigration in order to reduce housing demand and
make housing more affordable?
Professor Nickell: Well, that is politics, is
Q56 Lord Skidelsky:
Everyone has got their opinion.
Professor Nickell: My answer to that question
would be no, which does not mean to say I think we should not
curtail immigration, but you do not solve the housing problem
by curtailing immigration. That is what I would say. So the argument
is straightforward. It follows from my previous arguments, and
at the National Housing Planning Advice Unit we have been making
projections of where house prices are going, and at current rates
of house building we reckon that the house prices to income ratio
(that is at current rates of house building) would rise from seven,
where it is now, to around 10 after 20 years. That rise is in
part generated by the projected levels of in-migration, but making
the same sort of calculations as I was telling you about before,
supposing you reduced net migration to zero for the next 20 years,
I reckon the house price to income ratio would still rise but
to somewhat over nine rather than to 10, so we would still need
to build lots of new homes if we wish to stabilise affordability
and so there would still be a housing market challenge. Of course,
you may wish to curtail in-migration in other ways, for other
reasons, but, as I say, that would not change the fundamental
housing market challenge.
Q57 Lord Macdonald of Tradeston:
Just another aspect on housing, I assume there is a strong demand
from the indigenous population for social housing?
Professor Nickell: Yes.
Q58 Lord Macdonald of Tradeston:
I think it is probably unlikely that will be met by new building
but you are saying there are some groups that are disproportionately
involved in social housing, which obviously creates tensions in
Professor Nickell: Yes.
Q59 Lord Macdonald of Tradeston:
Can you see a way of squaring that to give perhaps more opportunities
for the indigenous population to get into social housing?
Professor Nickell: The way I think about the
housing market is the following: the first thing you have to do
is if you have more housing in the market sector so that prices
in the market sector moderate, then, almost by definition, there
would be more people moving into the market sector and less pressure
on social housing. That is the first point. Secondly, to put it
bluntly, the reason why there is a demand for social housing is
because people do not have enough money. Those groups of migrants
who disproportionately make use of social housing by and large
do so because they are poor, they do not earn enough money. It
is not specifically because they are migrants; it is because they
are individuals who do not earn enough money. You might ask why
certain groups of migrants have low levels of earnings and that
may be to do with their education, and so on and so forth, but
if you then wish to say when you are allocating social housing,
"We want to allocate social housing this way or that way,"
that is basically up to the people who allocate social housing.
If they feel that they have to give preference to certain types
of individuals, then they will do so. I am not sure that they
should but that is what they will do.