Examination of Witnesses (Questions 40-59)|
30 NOVEMBER 2006
Q40 Peter Viggers: How confident
do you feel about trade data? I am referring specifically to missing
trader intra-community fraud. The swings there have been a quite
Mr King: They have.
Q41 Peter Viggers: Some 13 billion
down to 4 billion. How confident are you as to the data?
Mr King: We are very unsure as
to what the true picture is. Some years ago, when people were
studying the black economy, the fairly obvious point was made
that, by definition, the black economy was rather difficult to
measure by the official statisticians. Here we have an extraordinary
position in which we have an official estimate of fraud. I am
not in a position to judge the extent of fraud, but the numbers
are very big, and so they do matter in terms of judging the underlying
trade picture. It is also the case that there are real measurement
problems in trying to gauge the value of international trade.
It is no longer quite as simple as weighing the containers that
went out through the ports or weighing the containers that came
in. Trade in services is now very large, and, if you look at the
data on a quarterly basis, you can see very clearly that the estimates
have a complete zigzag pattern. It is up one, it is down the next,
a classic sign of measurement error in the data. Even in the six-monthly
seriesI used to not look at the quarterly data for trade
but just the six monthly figureseven there you see a zigzag
pattern, and so I am inclined to think that the data on trade
may be only valuable on a calendar year basis, not on a more frequent
horizon at present. I think the honest answer is that we really
do not know. I do not think you can blame the Office of National
Statistics (ONS) for this, they did not instigate the fraud, they
have to try and cope with the measurement consequences of it,
and it is extremely difficult.
Q42 Peter Viggers: Another (a Rumsfeld
expression) "known unknown" is immigration numbers?
Mr King: Yes.
Q43 Peter Viggers: Do you have any
comment on the uncertainty there?
Mr King: I have obviously commented
at length before and I am happy to answer the question today.
There is enormous uncertainty about it. We do not have in the
UK any particularly accurate method of measuring migration, either
gross or net. It is the net migration numbers that matter, and
the gross flows in and out are very large. The only basis we have
really, that is used by the ONS, is the International Passenger
Survey (IPS). There are other sources of data, and I think we
are trying to encourage everyone interested in this area to try
and look at these other sources of data to give a cross-check,
and I think the Civil Aviation Authority Data on the number of
flights in and out will prove very helpful, but it is not a magic
answer here. The IPS is just one example which I think illustrates
the problem. Obviously, we do not know how many people have come
on a net basis from eastern Europe (the accession countries) to
work in the UK. If you look at the International Passenger Survey
for the answer, the numbers are surprisingly small. I think the
official estimate is below 100,000 a year. There are some reasons
for that. One is that in the International Passenger Survey to
be counted as a migrant you have to say that you intend to stay
for at least a year. There are many people who may come to do
seasonal work who may go back and return again, so there could
be a lot of people coming all the time to do seasonal work, or
there could be people who intend to come in for only three months
and end up staying longer, or there could be people who think
it is pretty dangerous to say they intend to stay for more than
a year. What is the incentive to say that? That is one difficulty.
Another one is that the change in the pattern of arrivals has
been pretty dramatic, and that is very hard for the ONS to catch
up with. The example, I think, which illustrates it most clearly
is the change in movements between the UK and Poland. In 2003,
I think there were 516,000 passenger journeys between the UK and
Poland. That is both in and out. Five hundred and five thousand
of thosethat is almost all of themwere to Gatwick,
Heathrow or Manchester. Over the next two years the number of
passenger journeys between the UK and Poland went from 516,000
to about 1.8 million. Almost all of that increase was in airports
other than Heathrow, Gatwick, and Manchester. So, from a figure
of 11,000 in 2003, there were more than a million passenger journeys
between the UK and Poland only two years later, all in airports
outside Heathrow, Gatwick and Manchester. Why does this matter?
Because most of the people handing out the questionnaires for
the International Passenger Survey were at Heathrow, Gatwick and
Manchester, so they missed a lot of these people. It is very hard
for them to anticipate the increase in traffic, they cannot hire
so many people to go round, and, in fact, the number of migrants
who actually said in the International Passenger Survey, "Yes,
I am a migrant coming into the UK", was 79. The numbers are
tiny in the survey, and this survey was never designed to measure
migration, it was designed to learn more about tourism and business
travel; and it is a perfectly good survey for understanding tourism
and business travel, but it was not designed, and you should not
expect it to come up with, particular answers for net migration
between the UK and the rest of the world.
Q44 Peter Viggers: Leaving aside
the sociological and economic effects of these two great big areas
of uncertainty, purely to do your job, as the people whose duty
it is to measure these things, is there more that should be done
get a clearer grasp on these two huge areas of uncertainty?
Mr King: Yes, and, obviously,
each year we state our priorities to the ONS within the framework
of the relationship we have with them, and for some time we have
been drawing attention to migration as a high priority area. But
the demands on the ONS, relative to the resources which they have,
have grown, and that is not a question for us, that is a question
Q45 Chairman: Is it affecting our
analysis of the statistics? I am reminded of the inquiry we did
into the scotch whisky industry, and Customs came back with ONS
figures saying that the level of estimates vary between 150 million
and 1,150 million, and so the opportunity for doing any rational
analysis on that was right out of the window.
Mr King: There is no easy answer
to any of this, because in a free market economy you always have
to use samples to obtain data. We can certainly do our job. I
am not saying our job is impossible; it is not. It is made more
difficult by the statistical uncertainties.
Q46 Chairman: You if get figures
like 150 million and 1,150 million, Governor, you cannot work
Mr King: What we are trying to
do on the migration front is to look at figures from all other
sourcesso the Workers' Registration Scheme numbers, the
number of National Insurance numbers which are granted for employees,
these data from the Civil Aviation Authorityto see whether
we can use them as a cross-check to get a feel for the plausibility
and the orders of magnitude. But these are developments which
will always occur in an economy. I think what I would like to
stress more than anything else is that, if anyone thinks that
setting interest rates is like a text-book case in which there
is a well-known model of the economy and seven equations, we know
exactly what the equations are and we know the parameters, we
know exactly what the data are, we feed it in, turn the handle,
out comes the interest rate, it is very, very different from that.
Q47 Chairman: You have educated us
to know that we do not go along those lines, Governor, so do not
worry about that?
Mr King: Thank you, Chairman.
Q48 Chairman: Professor Blanchflower.
Professor Blanchflower: Obviously
we have had difficulties understanding how many people flow in,
and the passenger surveys and so on are indicative to some degree
of that, and we have the work of registration schemes and the
national insurance numbers, but we do not have any really good
estimates of the number of people who have flowed out. We can
count the number of people who have a national insurance number,
but are they still here at any moment in time? We really do not
know that. Not only do we not have a very good measure of the
inflow, we do not have a great measure of how many people have
left or come back or any sense of how long workers here intend
to stay for, so we have difficulties both on the inflow side and
on the outflow side too. We do not really have any surveys designed
to help us in knowing how many people have left.
Q49 Chairman: You are making a good
case here for identity cards. Do you want to help the Government?
Mr King: No, I think this adds
up to a good case for a properly resourced national census the
next time round.
Q50 Mr Breed: I have a quick question,
Chairman, and it may be slightly naive, but, on the back of what
you were saying to Mr Viggers in respect of housing costs and
everything else, is the CPI now an accurate reflection of inflationary
pressures in the country?
Mr King: Yes, it is. I think there
has been a lot of nonsense written on this, to be honest. Perhaps
the first thing to say is that if we were each to calculate our
own individual inflation rate around this table then we would
come up with quite different answers because the patterns of our
spending will vary enormously.
Q51 Mr Breed: Certainly pensioners
Mr King: Pensioners or people
on low incomes. The inflation rate for particular groups can differ
and there are indeed separate indices for precisely that reason,
to be used when setting benefits and so on. Our concern in setting
monetary policy is to know what is the average inflation rate
across the economy. From that point of view, the Consumer Prices
Index (CPI) is an extremely well constructed and well-researched
index. This is an area in which there is not anywhere near the
sort of uncertainty that we have just been talking about. There
are thousands of inspectors who go out on a particular day every
month to collect prices in different outlets all around the country
and those prices are returned and they are weighted together by
the expenditure pattern of households, on average, collected for
very detailed household expenditure surveys. There is still a
choice about the conceptual structure of the index. The CPI does
not include owner-occupied housingand you can argue whether
that is a good or a bad thing: those are conceptual issuesbut,
once you have defined what it is you want to measure, the CPI
is an accurate measure.
Chairman: Thank you very much. That is
very helpful for the Government because we do get a number of
submissions on the CPI and various criticisms.
Q52 Jim Cousins: One of the difficulties
we are, I suppose, struggling with here is that change may not
be driven by average positions. It may be driven by more extreme
positions. I wonder if I could ask Professor Bean on this. We
have just been talking about the impact of immigration. This has
come at a time when costs have been rising quite steeply in the
economy and there is a whole section of the Inflation Report that
is set to looking at the implications for labour costs in this.
I am quite confused by what I am trying to be told in the Inflation
Report, because, on the one hand, it is being suggested that take-home
pay has not kept up with the increase in energy and import prices,
and that would certainly lead you to think that the immigration
factors have had an impact on take-home pay, and, on the other
hand, the costs to employers of employing workers have risen quite
steeply and it looks as though the impact of energy and import
prices has been born by employers rather than by employees. Have
I worked this out properly?
Mr Bean: Yes. Essentiallyand
this cannot really be avoidedthe cost of higher energy
has to end up being borne by the labour force, by workers. Over
the past couple of years, when energy prices have risen markedly,
we have seen that, although pay growth has remained subdued, in
a sense it has not been subdued enough. So real take-home pay
for households has been growing more slowly. It should be said
it is not just energy prices here, it is also effective tax rates
that have a role to play. But the post-tax real purchasing power
of the wage has grown relatively slowly (there has been some pick-up
in the last year but that has been relatively mild). But the growth
has not been subdued enough because the cost of labour for firms
(that is, the labour cost relative to the price of firms' output)
has continued to grow quite rapidly, and more rapidly, it appears,
than underlying productivity. Effectively, the cost of employing
workers has been rising too fast and that is something that would
be a discouragement to employment going forward. It is important
for us that we continue to see subdued pay growth going forward
over next year to avoid, essentially, labour pricing itself out
Q53 Jim Cousins: Let me try to understand
what you have just been saying. You appear to have told us that
we have been through a period in which immigration into the country
seems to have been quite high; energy prices, tax rates, import
prices have risen quite steeply; and real take-home pay, as a
result of all of these factors, has fallen. You are telling us
that it has not fallen enough.
Mr Bean: Correct.
Q54 Jim Cousins: Is that the view
of other members of the Monetary Policy Committee? Professor Blanchflower,
what about you?
Professor Blanchflower: There
is a question, in some sense, about to what extent real pay has
fallen, in the sense that many of our measures exclude about 27%
of all the workers. Workers in small firms are excluded from our
aggregate pay measures; self-employed earnings are excluded. In
some sense, our measure of what has happened to pay takes out
the most flexible, so I have the somewhat slightly different view
that these numbers probably even overestimate the extent to which
there has been wage growth. There has been, to some degree, more
flexibility downwards than perhaps others see, partly because
our wage series exclude those people whose pay is most flexible.
I think there is an issue at that end. There are concerns there
about what really has happened to real pay. I think it has probably
been more flexible downwards perhaps than others think.
Q55 Jim Cousins: Professor Besley?
Professor Besley: I hesitate to
answer because the term "enough" here is a somewhat
loaded term and I want to achieve some clarity on that. The issue
is: has it fallen enough so as not to lead to a significant reduction
in the number of workers taken on by firms? That is the sense
of enough. It is not a normative judgment about the right level
of wages in the economy. In that sense, I would agree with what
Charlie has said and the announcement in the Inflation Report
which says we have not observed real wages falling enough so that
levels of employment have not suffered as a consequence. I think
it is part of the explanationmaybe only partof why
we have observed a rise in unemployment in the recent past.
Q56 Jim Cousins: I find that very
interesting. The next session of the Inflation Report goes on
to discuss something equally interesting which is inflation expectations,
on the one hand, compared with inflation expectation volatility.
I am not clear about which of these two things you see as being
the more important. What is going to affect the labour market
more, a general rise in inflation expectations or a switch-back
Mr Bean: It is certainly not the
volatility bit. It is inflation expectations. That is the thing
we are concerned about, and, in particular, were we to see a substantial
rise in inflation expectations and that to be associated with
an increase in nominal pay growth, then that would obviously tend
to lead to inflation turning out higher. Our concern is that inflation
expectations stay anchored near the target. We have seen our measures
of household inflation expectationsand it should be stressed
that these are not particularly good measures: we have to rely
on various surveys, including the survey that the Bank carries
outbounce around a bit during the course of this year.
We do not understand very well what drives inflation expectations.
But it appears to be the case through this year that inflation
expectations moved up noticeably in the early part of the year
at about the time when there was a lot of press coverage of the
increase in gas and electricity bills. It eased back during the
summer and then there was another movement up, again at around
about the time that there was a lot of press coverage about higher
utility bills and other aspects of inflation, including all this
stuff about inflation as measured by the CPI not reflecting true
inflation. The unfortunate thing is that these movements in inflation
expectations have not been tied that closely to what has actually
been happening to the measures of inflation. Because of that lack
of a clear link between expectations and the actual measureinstead
they appear to be linked to perceptions of inflationit
makes it harder for us to form a view about how expectations are
likely to evolve going forward. But it is crucially important
to us that those expectations remain anchored reasonably close
to the target.
Q57 Jim Cousins: Governor, if it
is the case that real take-home pay, though it has fallen, may
not have fallen enough, and if that also comes alongside a situation
in which inflation expectations may be rising, then, in order
for you to patrol effectively the distant frontiers of the great
stability (to use a phrase that the Deputy Governor used in one
of her recent speeches, in a very interesting way) are you not
going to have to start expressing views about such matters as
the level of energy prices or the fact that 4,200 people in the
City of London this year are going to get bonuses of more than
£1 million? Because, in the situations we are dealing with,
these could have a real impact on how people behave in the real
Mr King: The way in which we can
influence how people behave is in terms of stressing our commitment
to meeting the inflation target in the medium term; hence, as
Charlie said, meeting the objective of having inflation expectations
close to the target, and that means low volatility of inflation
expectations. I do not think the Bank should be in the business
of commenting on individual prices or wages, because there is
a great deal of change within the economy and we are concerned
with the average as a whole. We will simply behave in such a way
as to react to the changes in the economy in order to move interest
rates to keep inflation on track to meet the target. I do not
think our job should be to give sermons about who should be paid
Q58 Jim Cousins: Governor, if the
problem of maintaining the great stability is not in the average
but it is in the extreme, the extremes of 4,200 people in the
City of London getting bonuses of £1 million, real take-home
pay having fallen but not having fallen enough, are you not going
to have to strike a position about some of these things?
Mr King: I do not see why commenting
on 4,200 people is germane to our particular challenge of keeping
inflation on track to meet the target. It has particularly struck
us that so far wage increases have been pretty muted. That has
led not to a fall in real take-home pay but to much slower growth
in real take-home pay, but, as Professor Besley said, that has
not been sufficient to prevent there being some incentive to firms
to cut back on employment and that may have led to some pick-up
in unemployment. We are concerned with trying to judge what will
happen in future and we are pointing out very clearly that we
intend to keep inflation close to the target of 2%, and all of
those involved in bargaining for wages, whether on the side of
employees or employers, should take that into account.
Q59 Mr Gauke: I am relieved that
incomes policy is not going to be part of the fix.
Mr King: We are certainly not
in the business of doing it.