§ 12.47 a.m.
§ Mr. John Sutcliffe (Middlesbrough, West)My right hon. Friend the Secretary of State for Employment has had good news today. Despite that news, it is true that a high level of unemployment persists in face of massive reflation. It is a puzzling phenomenon.
I am grateful to have the opportunity in this debate to argue the need for better information on one of our greatest problems. We are spending vast sums in trying to cure unemployment, but we spend the minimum on its diagnosis. Our existing machinery for collecting the information has been likened in usefulness to looking down the wrong end of a telescope. I am worried lest we are squandering valuable time when we should be devising and implementing techniques of information-gathering far superior to any we now have.
Existing unemployment statistics, far from being a reliable measure of the economy overall, obscure the true nature of the problem. They are neither soundly-based nor sufficiently detailed. There is the so-called percentage of unemployment issued monthly by my right hon. Friend's Department. We learn from it that this 638 September 3.9 per cent. were unemployed. Naturally, we conclude that 3.9 per cent. of those actually looking for jobs could not find them. But the percentage tells us nothing of the sort. What it measures is the number registered with employment exchanges at a particular date as a proportion of the total number of employees both employed and unemployed.
Neither component of this measure is what we want. It is misleading. As a guide to the state of the economy or to unemployment it is unreliable. The number registered does not accurately represent the number seeking work. Many who want jobs do not qualify for unemployment benefit and they do not register—for example, recent entrants to the labour force, those discharged for misconduct, those whose entitlement is already exhausted, and those who have not paid national insurance contributions at the full rate, which goes for many married women. There are others who may not choose to collect benefit.
The record of employment exchange placement is another factor. Exchanges have been responsible for, perhaps, a fifth of placements overall and a mere 6 per cent. of executive and professional postings. Given this record, there has not been in the past any great incentive to sign on. Clearly the register excludes people who are seeking work, but it is misleading in another way. It includes many who are not seeking work. The Department makes some allowance for this by omitting school leavers and the temporarily stopped from its category of "wholly unemployed", but far more people than this are in who should be out. I refer, for example, to occupational pensioners and those nearing retirement. These people are unlikely to be looking for jobs. Then there are those who simply abuse the system, who collect benefit while simultaneously holding down a job or who prefer living on the dole to working for a living.
Even if the number seeking work were available, expressing it, as we do, as a percentage of the total number of employees would still be misleading. Employees are only a part of the working population and an index ignoring the armed forces, employers and the increasing numbers of self-employed cannot hope to be accurate. It will systematically over-state the percentage 639 of unemployment. In 1971 the wholly unemployed constituted 3.3 per cent. of all employees but only 3.1 per cent. of the working population. In each of the four previous years there was a similar difference.
We must revise the present system, not just in order to know what percentage of the labour force is actually seeking work, but to know much more than that. We need to know how unemployment breaks down in terms of occupation, duration, location, earnings, age, skills, and the potential to acquire new skills. Is it voluntary or involuntary, long-term or short-term? Short-term unemployment constitutes about 40 per cent. of the total. This is redeployment. It is to be treated differently from long-term unemployment. Those out of work for less than eight weeks scarcely pose a problem, certainly not one calling for reflation. Similarly, those out of work because of disablement require special sheltered jobs, not fiscal measures. Occupational pensioners require no action: they should not be classified with the unemployed. In contrast, young people in the first three years after leaving school need every help and the more we know about their problems the better.
I say this with real feeling. In the Northern Region last month there were over 12,000 boys and girls out of work. This is an appalling state of affairs. I told my right hon. Friend the Secretary of State for Employment in a letter the other day that it calls for emergency action.
We need to improve both the sources and kinds of information. We must ask more and better questions. For example: what is the job seeker's level of skill and education? This would help identify under-employment as well as unemployment. Has the job-hunter got dependants? Is he the breadwinner? What effort has been made to find employment? The data collected need to be thoroughly cross-analysed. We should not just seek to know what the percentage of the unemployed is in each occupation, but what percentage of each occupation is unemployed.
We cannot learn these things from the employment exchange register in its pre- 640 sent form. The register data could be made more accurate by separating the employment from the social welfare functions, by computerising and by further cross-analysing data. Whatever measures are taken to improve information from the register, surveys are needed to supplement the register as a source of data. Surveys are a well-established method of doing this in developed countries. They are regularly carried out in the United States, Canada, Japan, Germany and Italy. The National Institute Review estimated that whereas the published rate of unemployment for all workers in 1970 was 2.6 per cent., a household survey would have revealed a figure of 3.9 per cent.
My hon. Friend will be aware that the 1966 census actually revealed a colossal discrepancy. It discovered 186,000 more people job hunting than the register showed. Clearly a comprehensive survey like the census is the most effective way of gaining information. The analysis of the 1971 census will no doubt confirm this. Regular household surveys have an equally valuable räle in between censuses. We would have to resist the temptation to base samples on national insurance numbers. Such a system would leave out of account those seeking but unable to enter the labour force. We need information about jobs going wanting as well as those wanting jobs. Household counts should be supplemented by surveys of employers, commercial agencies and newspaper advertisements. Information about vacancies is woefully inadequate. At most 25 per cent. of jobs available are notified to employment exchanges.
All this sounds expensive but what is the real cost compared with the cost in human misery and wasted output resulting from a continuous high level of unemployment such as we have in the Northern Region? Sample surveys in the Northern Region carried out under the auspices of a professional body such as the Central Statistical Office should cost no more than one-tenth of the cost of the census in this region. Nationally every 1 per cent. of unemployment represents a loss in potential output of goods and services of about £350 million a year. The development areas are the obvious places for pilot projects. I suggest starting this winter.
641 All of us warmly welcome the downward trend revealed by the figures published yesterday but in the North it is a trend that starts from an unemployment rate of 7 per cent. in September compared with a national rate of 3.9 per cent. We can never be complacent about such a disparity. Present statistics may serve to demonstrate trends, but they mainly serve as a stick with which to beat the Government—happily, though, not this month.
I should like information so improved that it will make possible an effective manpower policy and, more purposeful training and retraining of our labour force to do work for which there is a known or projected demand. There are said to be 2,843 surplus clerical workers in the Northern Region. This information is useful. It tells us what we need to do. It tells the Government of the need to disperse Civil Service offices and to encourage business headquarters to set up in the North. Teesside, where my constituency is, manages to combine higher-than-average unemployment with a bottleneck in certain skills. We must identify in order to train and similarly to use any known surplus of skills to attract new industry.
We need urgently to improve information in order to get the maximum benefit from the Common Market assistance which will shortly be available to us. The European Social Fund gives financial aid for vocational retraining of unemployed or under-employed workers and for resettling workers who have to move to get new jobs, helping workers temporarily stopped while a firm changes its basic activities. Similarly, the ECSC assists in retraining and resettling redundant people in the coal and steel industries. Assistance given under these schemes amounted to 300 million dollars to the end of 1970, and this is expected to increase.
I am sure that my hon. Friend will not disagree with me that diagnosis must come before cure. In order to diagnose we must first be informed. Having listened so patiently to my arguments, I hope that he will tell me that he intends to secure better information and that he will make a start this winter by initiating pilot surveys.
§ 1.3 a.m.
§ The Under-Secretary of State for Employment (Mr. Dudley Smith)My hon. Friend the Member for Middles-brough, West (Mr. Sutcliffe) was right to talk about the good news we have had today about the latest very encouraging unemployment figures. While we must not read too much into one month's figures, there is cause for satisfaction that the Government's measures in this direction are beginning to work.
My hon. Friend will appreciate that the actual creation of employment in development areas is the responsibility of my right hon. Friend the Secretary of State for Trade and Industry, through the revised and increased incentives now offered in the Industry Act, 1972; and, as he will know, very substantial changes of organisation have been made to ensure rapid and more flexible implementation of these policies. We for our part in the Department of Employment have a responsibility, as my hon. Friend points out, to see that these new policies are supported appropriately by adequate information, and we do not underestimate the difficulties.
That means, primarily, information as to the kinds of manpower, including female labour reserves that are available in different places; to levels of earnings; so far as possible, to the intentions of employers and to local trends in employment and unemployment which point up those areas on which attention needs to be focussed.
The question my hon. Friend has raised amounts, in effect to asking what forms of local labour market intelligence we need to operate regional policy and what is the best way of collecting it—whether by way of household surveys or by other means. This is naturally a question which has been occupying our minds very much. Of course, we have always been active in this field. It would be wrong to underestimate, as I know is sometimes done, the wealth of knowledge of our local managers of their own areas, arising from their day to day operations—combined with the indicators we can give them of local unemployment, employment, activity rates, and so on.
But I do not propose to dwell on this aspect, because the discussion tonight is not about the effectiveness of what is 643 done, but about what more can be done and how, and whether through household surveys. Household surveys are generally most useful for providing national figures where one is not looking for very precise information about sub-groups.
If one is seeking to identify sub-groups, for example, unemployed people or people with particular skills or backgrounds, then of course the size of the sample required has to be very large, or they will not contain a sufficient number of the people concerned. Past experience shows, too, that questions requiring very precise answers, such for example as the level and type of skill of the householder, are frequently not properly answered and require a particularly high level of interviewing, even to get approximately satisfactory replies.
None of us, after all, is always particularly modest about our achievements, and if we are asked about them, some of us can be pretty reticent about our motives. That presents considerable difficulties for those conducting surveys.
Consequently, surveys seeking the kinds of precise answer that would be useful to us in the context my hon. Friend has in mind are less useful and much more demanding in terms of resources than might appear at first sight. If we ask ourselves how to apply our limited resources to best effect most quickly to get a continuous and accurate picture of changing local labour market requirements, particularly in development areas and of labour reserves, I do not think that we would decide to go about it this way.
Of course, that does not at all rule out the use of interviewing and household survey techniques to get at particular aspects of the labour market where the changes that are taking place are slower to operate. We sponsor a good deal of research of this kind. It includes research into the absorption of labour after major redundances—a subject of particular concern in development areas—and the characteristics of specific groups of unemployed people, such for example as the long-term unemployed. This kind of research is not a substitute for better statistical knowledge or better labour market knowledge. It is designed to supplement them both.
644 I should now like to say something about the usefulness of household survey techniques for more general statistical purposes. Since the end of the Second World War, household surveys have been increasingly used in many countries as a means of providing information about changes in the economic and social characteristics of the population. In some cases, these surveys are referred to as labour force surveys, when they have concentrated primarily on providing information about the labour force, and its main components, and on employment and unemployment, and how these relate to numbers in the total population.
Current public interest and concern with the higher levels of unemployment here have prompted a number of commentators, and my hon. Friend, to criticise our present unemployment statistics. The critics suggest that better quality figures could be obtained from surveys addressed to representative samples of households, instead of relying, as at present, on those registering at employment exchanges.
I do not propose to consider in depth because of the lack of time, the various criticisms which have been made about the present unemployment statistics for the United Kingdom. The make-up of these statistics and their presentation have just been considered by an Inter-Departmental Working Party of officials and their most useful and important report is now being considered by Ministers.
I expect that it will probably be published in a few weeks' time, and I think that it would be better to defer further discussion of the present unemployment statistics, and of their presentation, until then and I am sure my hon. Friend would agree.
I must stress that it is particularly important that statistics of employment and unemployment should continue to be obtained for local areas, because of their vital role in determining regional policies and other planning requirements.
This means that information from labour force surveys could be used to supplement, but not to replace existing unemployment data. It is inevitable that such information would conflict, to some extent, with trends shown by the traditional figures, but these disadvantages 645 would be more than offset if household surveys contributed significantly to out better understanding of labour force trends, for example, about the relative movements of employment and unemployment and how both of these are related to changes in the total population.
The labour force surveys might also provide additional useful information on other relationships, for example, about the social and economic characteristics of those seeking work which my hon. Friend referred to. Additional questions aimed at providing information of topical interest can also be included. At the present time, when the Government are launching new schemes for training opportunities, it would be useful to include questions about training and occupation. Some information of the kind provided by labour force surveys becomes available periodically from the more comprehensive censuses of population, and we look forward to studying the detailed information which is now beginning to emerge from the most important 1971 Census of Population.
Another recent development is the General Household Survey, which is a small continuous household survey addressed to about 15,000 households a year. The relatively small size of this survey means that it can only contribute a limited amount of information about labour force trends. Its primary purpose is to explore on a continuous basis some of the changing relationships over a very wide variety of topics, covering employment, education, health, housing and so on.
Member states of the European Economic Community have periodically participated in labour force surveys, which have been designed to provide comparable information about labour force trends for the community as a whole. The next EEC Labour Force Survey will be carried out in 1973 and the United Kingdom has been invited to participate. The survey in the United Kingdom would be expected to cover about 100,000 households and, if we participated, it would offer an excellent opportunity to assess the value of information obtained from the household surveys. Nevertheless, the survey is expensive and the net 646 cost to the United Kingdom is expected to be of the order of £500,000. I must tell my hon. Friend that the Government have not yet decided whether or not the United Kingdom should participate. My hon. Friend's comments tonight are valuable and will be borne in mind in this connection.
The question still remains, if household surveys are of limited usefulness for local labour market intelligence work, which way do we go to get information relevant to these, and to regional policies? From last December, my Department has been conducting experiments in nine areas on how the collection and use of local labour market information can best be developed in the interests of work people. Emphasis has been placed on finding out direct from employers what developments they expect which will affect manpower demand in the local market. The object of the experiment is to see whether we can collate information of this kind so as to make it available to others, with due regard for confidentiality and in a way in which all concerned with manpower planning will find useful. If the experiments this year are judged successful, they will be extended to other areas. We shall hope to be able to build up a two-way exchange of information on local developments and perhaps also to anticipate wider trends.
This therefore is the way we are already going. We all hope that it will help not only to impart greater knowledge and realism into regional development but be of wider significance in forming national labour policy. We are by no means complacent about this question. The results are important and significant in getting the right type of data. We are making active plans for the whole reformation of the employment service to improve many aspects of that service, not least of which is job notification which is very important.
I am grateful to my hon. Friend for his comments, which we shall study. I assure my hon. Friend that we will spare no effort to continue our efforts to reduce the general level of unemployment.
§ Question put and agreed to.
§ Adjourned accordingly at fifteen minutes past One o'clock.