There are, on the other hand, the three coefficients mentioned, V (the scale effect), the Es (the elasticities of both foreign and import demand) and to a lesser extent s (the elasticity of substitution), which do bear more directly on the difference between the immigrant time-path and the non-immigrant time-path. Because of the greater sensitivity of the results to these three coefficients, one must take the precaution of experimenting with a range of values for each of them, as indeed has been done in our calculations.

The figure of 500,000 immigrants per annum over a 30-year period may appear high to some people even if it were supposed that the U.K. resumed its traditional "open-door" policy at least for Commonwealth immigrants. But the figure in itself is of no importance. The only reason for choosing it is to ensure that the changes wrought in our indices A, B and C are all large enough to be perceptible. The assumption of a rate of inflow below 500,000 per annum would imply figures (roughly) proportionally smaller than those in the Table, and vice versa.

At all events, the estimates for B and C shown in Table 2 are significant only inasmuch as they are surprisingly small. Those who maintain that immigration imposes large economic losses on the country cannot derive much support from this sort of analysis – not unless they anticipate immigration on a scale very much larger than the 500,000 per annum assumed here. As for those who anticipate economic advantages from net immigration, a case could be made out in terms of the B and C indices only if evidence could be produced of large economics of scale for the country as a whole – that is, of a V larger than 1.2. If, however, the conventional view is allowed to prevail, and for the economy as a whole constant returns to scale are assumed, one must anticipate some decline in over-all real income per capita (indigenous plus immigrant), and some per capita net loss for the indigenous population alone. As suggested above, however, neither is large when taken as a proportion of income per capita or of aggregate income respectively. For an annual migrant inflow of 500,000, for instance, the decline in per capita real income, B or C, does not exceed 6% after 30 years in any of the three cases. This limited decline in per capita real income, B or C, can be attributed in the main to the built-in neutral technological progress that is a feature of our model.

Only for A, the index of distribution, does the outcome look somewhat more sombre. In all three cases immigration makes the resulting income-distribution distinctly more regressive. All three cases, however, assume a s, of unity. If the s, instead were equal to 2, the consequent rise of the A index by 11% over the 30-year period – rentals rising 11% more than wages compared with the position in the absence of net immigration – could be borne with, since real wages would in any case be rising at an average annual rate of about 2 %. Per contra, if s were 0.5, there would be a more than 50% rise in the A index over that period which, for the same immigrant numbers, could effectively prevent real wages rising over time, or very nearly.

Is a value of 0.5 for s at all realistic? There is a tendency today to be impressed with the fixity of proportions in any given state of technology, so that such a value would not be thought implausible. But even if we accept that the proportion of labour and capital is relatively impervious to changes in the prices of labour and capital, we could invoke the opportunities for product-substitution to impart flexibility and limit the relative decline in wages." As suggested above, however, we know practically nothing of the actual size of the consumption effects to be anticipated, and we have no option but to conclude, rather lamely, that the regressive distributional effects of a s of 0.5 cannot be ruled out on grounds of plausibility.

Summary

In a long-run aggregative model of the economy in which we ignore all "temporary" dislocations, all inflationary effects and all external diseconomies, we may plot a movement of three indices, A, B and C, over a 30-year period by comparing an immigrant with a non-immigrant time-path using a number of arbitrary but plausible assumptions for the relatively insensitive parameters while experimenting with the more sensitive ones. Attributing fairly conventional values to the sensitive parameters as in Cases (i) and (2) - s of 1, V of 1, and Es of 1.5 or 2.5 – Table 2 reveals a fairly pronounced regressive distributional effect over time, a decline in the B index of per capita income over 30 years of less than £3312 per annum, and a decline in the C index of per capita income of less than £55 in the 30th year. A large enough increase in the economies of scale for the country as a whole – a V of 1.2 or more – would, however, raise both the B and C index of per capita income to about £62 and £18 per annum respectively. But a value of 1.2 for V would not be regarded by economists as realistic for the country as a whole, and pending evidence contrary to the prevailing belief one may conclude tentatively that an economic case for large-scale immigration, at least one based on indices which would be widely acceptable, is not proven. On the other hand, on the basis of the same indices, the economic case against immigration – except on a scale much larger than that conceived in the analysis – is not compelling save perhaps in respect of distributional effects.13

III. Immigration & "Essential" Services

The popular belief that Commonwealth immigration has helped to overcome shortages in particular service industries may now be examined in an informal way.

It is not always certain, however, that Commonwealth immigrants' entry to an industry suffering from a labour shortage invariably acts to reduce the shortage. If there is among some of the indigenous workers a dislike of working with some kinds of immigrants, or if some stigma comes to be attached to occupations that employ a large proportion of coloured workers, the entry of coloured workers into an occupation can prolong or aggravate the initial shortage of labour in so far as it causes some of the existing workers to leave and in so far as it deters those of the indigenous population who might otherwise have entered. In such circumstances the effectiveness of Commonwealth immigrants in remedying a shortage is reduced and could indeed be negative. The observation of a large proportion of immigrant workers in occupations most easily accessible to them is consistent on this hypothesis with very little increase in total numbers and possibly with a continuation of the shortage. Public transport and nursing are occupations that might well belong to this category.

Allowing, however, that on balance immigrant labour is eventually effective in relieving an initial shortage, is there any clear advantage in following a policy of admitting foreign labour into occupations that are short of labour for the time being rather than adopting the alternative policy of meeting the shortage from existing domestic resources?14

Consider first the allocative aspect. To a passenger depending upon a bus or train service, its maintenance at the same fare is understandably preferred to its withdrawal or to its continuance at a higher fare. But this is clearly a partial view only. In the absence of immigrant labour, which may realise this outcome, this sectoral shortage would be remedied in part by a differential rise of wages in public transport. All intra-marginal workers in such an occupation would gain, and this gain must be set against the loss to passengers. A transfer of "real" income from indigenous passengers in favour of indigenous (intra-marginal) workers is the apparent outcome of the non-immigrant solution to the shortage.

But what of the allocative effect? If the shortage in any industry is remedied as effectually by an inflow of labour from abroad as by a release of labour from domestic sources, and if in either case labour moves until the value of its marginal social product is the same in all occupations, then there would be nothing to choose as between the immigrant and the non- immigrant solutions except for two things, one in favour, one against: (i) the non-immigrant solution may take longer, so prolonging the loss to the consumers of the service, and (ii) the immigrant solution may imply a lower level of welfare for the indigenous population as a whole as measured by the indices in Part II – though, as also indicated, such losses as a proportion of the relevant magnitudes are not large. Since these two considerations are opposed in their effects on the indigenous population, quantitative estimates would be necessary to determine the net result.",

Let us turn next to the long-run consequences of accepting a policy of admitting foreign labour to industries claiming to suffer from a labour shortage. In an advanced and fully- employed economy subject to continual fluctuations in the conditions of demand and supply, shortages in some sectors are sure to appear from time to time matched by surpluses in other sectors. The duration of such shortages, and surpluses, will depend inter alia on institutional factors (trade union influences and demarcation rules) and also on the active policies pursued by governments which bear on unemployment pay, retraining facilities, incentives to geographical and occupational mobility, monetary and fiscal management, and so on. A policy of encouraging immigration whenever a sectional shortage of labour occurs would, because of its manifest asymmetry,16 issue in a continued net inflow of foreign labour into the country having the broad effects discussed in Part II.17 However, any rule that sanctioned the admission of immigrant labour into any industry after the persistence of an unfilled vacancy beyond some agreed time-period might well lead to increasing friction between management and labour. For such a rule would clearly act to discourage any employer from negotiating wage increases and from providing facilities calculated to attract domestic labour if, by waiting a little longer, he can meet his requirements by immigrant labour at existing wages.

Finally, there is the broad question of the so-called "optimal" size of population – sometimes identified by economists with a population for which (within a static framework) average product per worker, or per capita, is highest. As indicated in Part I, however, there can be no "economic need" of a larger population or, more precisely, there are no clear economic advantages in Britain either in the short or the long run of a larger population – though landlords and businessmen favour a continued growth both for its immediate market-expanding effects and for the long-term distributional effects in their favour. If a population larger than that which would result from the growth of the existing domestic population is believed to be desirable on "non-economic" grounds-or desirable subject to some restriction on the rate of net imraigration – in full awareness of the initially inflationary effects and the long-term regressive distributional effects, then the economist has little to add.

Certainly the so-called optimal population is not an unambiguous economic concept: it can be defined in a number of ways, such as the population yielding the highest per capita income, or the population enjoying the highest over-all level of welfare per family, or that suffering the fewest adverse neighbourhood effects. If an optimum population has any affinity with the latter sort of definition then it is certainly relevant to observe that population per acre in Britain is one of the highest in the world. In particular, the area of England and Wales is today more densely packed with people than either Japan or Belgium. It has about twice the population density of Italy and four times that of France. India, Jamaica and Pakistan, from where the bulk of the post-war immigrants have arrived, have each less than half the number of people per acre than England and Wales. Holland alone can boast a country more densely populated than ours.

In an era as conscious as is our own of the impending "population explosion", any proposals for augmenting the already dense and growing population of this country by an influx of people from other lands can no longer count on ready acceptance.

IV. External Efects

Some consequences of large-scale immigration do not lend themselves so easily to measurement as those considered in Parts I and II, but they may be at least as important and a good deal more noticeable. A number of these effects are readily classified as external diseconomics. In the short run immigrants tend not only to settle in the existing conurbations but, within them, to concentrate their numbers in popular areas or districts, so manifestly aggravating an existing housing shortage and imposing additional burdens on the social services and possibly also on the public transport systems. Inevitably they reduce, for some years, the amenity of the neighbourhoods they settle in.

One can, of course, dismiss such external effects as teething troubles necessarily associated with the process of settling down. Keynes' dictum, that in the long run we are all dead, is relevant in this connection; for it is during the short period in which we live that the discomforts have to be borne with. Such net disutilities as are suffered by segments of the indigenous population are unambigiously a part of the cost of absorbing numbers of immigrants and wherever possible they should, as Pigou put it, be "brought into relation with the measuring rod of money".

External cffects, however, can be broadly or narrowly defined. On a broad definition any response, positive or negative, of any inhabitant of the host country to the entry of any or each of the immigrants qualifies as an external effect. On such a definition it is not necessary that the economist be able to identify any benefit or damage to an inhabitant arising from the entry of immigrants. The effect can be solely subjective – "pure prejudice", if we like. Without ever meeting a single immigrant the mere knowledge that immigrants, or immigrants of a certain type, are entering the country can add to or subtract from the satisfaction of any member of the host country. On the so-called Net Benefit criterion (which is the foundation of all allocative propositions in economics), unless the gains from immigration can be so distributed as to make every person included in the host population better off, the host country is not to be regarded as better off.

On this criterion, then, an economic improvement would require that, irrespective of his prejudices, the sums needed to compensate each member of the indigenous population suffering any discomfort whether "real" or "imaginary" from the entry of immigrants could be more than covered by the gains made by other members of the indigenous population plus the gains made by the immigrants.

A narrower definition of external effects, one I favour, would exclude all those responses to others' behaviour that cannot count on an almost unaminous approval from the society in question. Thus evidence of direct damage to a man's property, or health, or physical environment, or peace and quiet would in Britain, I think, be almost universally regarded as relevant to the issue. Dissatisfaction arising solely from private principles or prejudices would, however, not qualify as agenda on this narrower view.

If we accept the narrower definition of external effects, the ideal experiment is then to determine for each locality into which immigrants enter, or affect by the repercussions they generate, the minimum sum (reckoned either as a capital sum or annual payments) which, if received by the affected members of the indigenous population on condition of admitting into the localities in question a known number and type of immigrants, would in practice make the members indifferent as between receiving the immigrants and maintaining the status quo.

Needless to remark, no one has yet attempted to estimate the magnitude of these compensatory payments, and while this is no reason for failing to mention them, or for failing to dwell on them,19 it must be admitted that there is no firm basis even for a guess at the social costs involved.20 For the present, then, no more can be done than explicitly to acknowledge their incidence and potential significance, and to offer the obvious generalisation that they are likely to increase with the number of immigrants, their rate of arrival, their initial level of poverty and lack of sophistication, and with the degree of their concentration within already densely populated areas.

In conclusion one may hazard a prediction. With the inevitable extension of communications there will follow, among the economically under-privileged, an acute awareness of the increasing disparity between their standards of living and those of the ordinary workers in economically advanced countries. In the absence of government checks to immigration, the growing temptation to migrate to the few prosperous countries open to them would be strengthened by private shipping and airline companies which would find it profitable to encourage mass migration by offering cheap passenger rates and credit facilities. As it is, and even in the presence of government controls, one can reasonably anticipate a growth in the numbers of illegal immigrants into the wealthier countries, in particular into Britain and North America.

We are not, however, precluded by the above observation from making attempts to ameliorate the economic conditions of the poorer countries if we conceive it to be part of our moral duty to do so. Though some proponents of liberal dogma appear reluctant to concede the possibility, an economic policy may yet be acceptable to the nation without necessarily redounding to its material advantage. If, therefore, on moral grounds we wish to make some contribution to the well-being of the poorer countries we could make our contribution the more effective by undertaking a careful examination of the various methods of affording economic relief (such as removing trade barriers to their sales of products in this country) and by giving direct aid as an alternative to the policy of transferring some part of their growing populations to these already crowded islands.

Considerations of distributional justice would seem to favour exporting capital rather than importing populations. We should hardly regard it as fair to earmark the additional capital sent abroad – which would be an alternative to using it to equip immigrants entering Britain – to be distributed among those families that might otherwise have entered this country. Rather we should want our aid distributed within the poorer country according to some more acceptable principle of social priorities. As an alternative to the export of capital, the import into Britain of a small proportion of their populations is, then, a highly unsatisfactory method of distributing economic relief to poor countries. As a means of promotinge conomic advance in such countries this alternative is also less efficient on allocative grounds.

A Postscript on Commonwealth Immigration and Race Relations

Concerning the connection between immigration and race relations there are a number of elementary facts that can bear some brief emphasis.

First, parallels drawn by politicians between post-war Commonwealth immigration and immigration in the past of Huguenots, Jews, and Irishmen provide no guidance whatever to immigration policy today. The fact that particular immigrant groups entered this country in the past proves no more than that, notwithstanding the event, the country has survived. There is no evidence to suggest that in each and every instance the consequences for the host country were wholly, or on balance, beneficial. Certainly there is no foundation for the belief that under all circumstances immigration of any peoples in any numbers confers immediate or subsequent benefits on the host country. The conclusion seems unavoidable: granted that the interests of the host country are one of the considerations in determining immigration policy, the question of whether and which immigrants should be admitted is a legitimate subject for public debate.

Second, governments at least should be familiar with the historical observation that the reception accorded to foreigners settling in the host country varies, inter alia, with their numbers. Such is the imperfect state of man that there is apparently a limit to the number of immigrants entering a country beyond which any initial cordiality of the native population turns to resentment. That this tolerance limit for any given country depends upon a variety of factors is also indisputable. The more obvious of these are the wealth, the education, the independence and skills of the migrants, their ability to speak the language, their readiness to acclimatise themselves to their new social environment, and the degree of their dispersion over the country. Not the least in importance, alas, are their colour and physical characteristics, which may be alleged to operate on the popular mind at an irrational level.

If our enlightened rulers were aware of these facts, they must have thought it proper and decent to ignore them. For it is apparent now that we must have passed this tolerance limit at some time in the '50s-

Third, the transformation of Britain during the 1950Ss from a relatively stable and homogeneous nation to a multi-racial concern was not historically inevitable. Nor is it, in the increasingly race-conscious world of today, a manifestly desirable consummation. At all events the British public was not given the opportunity to debate the question. Through a tacit understanding between the leaders of our three parties to act like "enlightened" liberals and to put off for as long as possible any measures of control that might be construed as "racialist", and through the connivance of the Press, the country has simply floundered into a multi-racial society.

Fourth, it is not surprising that the people whose inaction was responsible for this gratuitous social transformation should now seek to persuade us of its inevitability, and of its desirability, and to soothe our apprehensions of the dangers we are courting by recourse to the language of "challenge". It is no more surprising that a large part of the public, belatedly discovering the transformation, feel that somehow they have been "taken for a ride" by the intellectuals. At no time can they recall being presented with a clear choice of whether or not to transform their country into a multi-racial, or multi-coloured, society. On an issue having momentous consequences for their way of life, their opinions were never sought and their wishes never consulted. Indeed, as indicated, the question was never explicitly formulated, and the range of economic and social implications of embarking on this venture was never openly debated.

It ought therefore to be put on record that the ordinary people of this country were, in this instance as in others, the hapless victims of liberal dogma and complacency. It may be "enlightened" or "humane" dogma, but it is dogma just the same. And if there is a lesson to be drawn from the sorry history of such events, it is that men today whose political instincts are guided by such dogma, and not by ruthless pragmatism and forethought, are unfit to grapple with problems in a world erupting with new forms of social strife, the consequence of over-swollen populations and of rising tides, everywhere, of unrealisable expectations.

Fifth, if we discount the hope of raising sums sufficiently large to tempt the large majority of Commonwealth citizens to return to their homelands, we shall have to resign ourselves to the fact that Britain is now, and will henceforth remain, a multi-racial society. Indeed, an effective population policy will be that much more difficult for the country to agree upon since there is an obvious inducement for non-white residents to increase their population faster than the indigenous inhabitants in order to increase their political leverage.

There is, then, everything to be said for our going to considerable expense and trouble to prevent the virus of American race conflict from infecting our shores. At the same time we can only hope that anxieties over racial issues will not, as they have done in the United States, reach such dimensions as to distract the public's attention from the complex of urgent social problems created by rapid urban and economic growth.

Sixth, the measures that are now being taken'in good faith have disadvantages that can bear thinking about. The establishment of a Race Relations Board and the enactment of special race legislation may, perhaps, seem justified in the immediate circumstances. But they also act continuously to remind us that there are elements of society in our midst for which the ordinary laws and institutions of the country do not suffice. What is more, such machinery as is put together will not be easy to dismantle. Owing to the self-justifying propensities innate in bureaucracies, we may inadvertently have initiated another "'growth industry" whose chief beneficiaries will be (perhaps otherwise unemployable) social scientists.

Press coverage of race incidents and alleged discrimination is unavoidable. But informing the public without inflaming any part of it is the ideal to be pursued. What is of equal concern is the recent flood of newspaper articles and television features calculated to shame the native population into better treatment of the newcomers.

It is doubtful whether this policy acts to promote racial harmony. Not only can it become a frightful bore: the repeated presentation of immigrants as innocent underdogs, whose plight is almost wholly attributable to callous treatment by the indigenous population, is worse than doing nothing. For it creates a hyper-consciousness about coloured people, an excessive caution in their presence, an over-considerateness and artificial humour in dealing with them that is productive only of repressed irritation on both sides. Even the most liberal White begins to jib at the endless tweaking of his conscience by journalists eager to find a vocation in exposing racial disharmonies. Such consequences are subversive of progress toward an easier and more open relationship which is a precondition of racial harmony.

It is not impossible, then, that less pressure from above, less condescension, less publicity, less musing and moralising by all of us, will-granted that the immigration phase is really at an end-better enable the new races to come together with the existing population and in time, and in their own way, discover their common humanity.