The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
FEDERAL EXPENDITURE AND ECONOMIC STABILITY GOVERNM ENT SPEN D IN G AND ECONOMIC S T A B IL IT Y W alter D. Fackler, assistant director of economic research, Chamber of Commerce of the United States I n t r o d u c t io n Economic stability is one of the goals of public policy which almost everyone espouses and about which few seem to agree. The term “sta bility” has a ring of high purpose. I t is a laudable and lauded objective—as long as it is left vaguely defined. As soon as discussion passes from pleasant generalities to what is meant by stability and the appropriate means for achieving it, controversy ensues. Dis agreement becomes more intense when policy measures are proposed or put into effect which are designed to promote stability. There are good reasons for this state of affairs. Economic sta bility as a policy goal in a changing world is not a simple concept. I t is rather, in the jargon of Washington, a “ball of wax” of remark able pliability. People do not want stable incomes; they want rising incomes. They want a stable general price level, but they also want higher prices for the products and services they sell and lower prices for the things they buy. They want technological progress, greater productivity, opportunity, and economic freedom of choice. They want normal market adjustments to allocate resources and direct production so as to obtain maximum output composed of the “right things” at minimum cost. But they are understandably vexed or alarmed when market adjustments affect them adversely, when they must move to another “line” or suffer permanently reduced income. I t turns out th at people do not want stability—at least not too much of it. They want security from the threat of unemployment, disagree able economic pressures, and income losses. In short, stability as a policy goal, beyond the vague injunction to public officials that they must learn how to keep us out of situations of mass unemploy ment and advanced inflation, is actually a mass (mess) of conflicting goals. Even when pragmatic agreement is reached for policy purposes on a reasonably acceptable concept of economic stability, major disagree ments are bound to arise over the means of achieving stability. Again, with good reason. Legitimate differences of opinion will exist as to the correct interpretation of economic events, and the relative impor tance of the many economic and noneconomic forces operating m a given situation. There are differences in value judgments as to importance of other policy goals, the proper role of government, the role of the individual, the kind of economic system which is desirable, 325 326 ECONOMIC GROWTH AND STABILITY and how different policy measures will affect the system in the long run. Moreover, since it is possible, at least theoretically, to achieve stability by various combinations of public measures, each individual group will plaintively or arrogantly demand th at combination of policies which will secure for it maximum economic advantage or least disadvantage. The first maxim of special pleading, as legis lators know so well, is to identify the particular interest with the general interest. This would be a strange world if it were otherwise. Because of both the shifty character of economic stability as a policy goal and the shifting sands of argumentation over how it should be promoted, only the naive or tyrannical will expect or de mand the impossible—spontaneous and harmonious agreement in the arena of public policy. But all is not lost. H aving stressed the inherent difficulties of the problem as a prophylaxis against dogmatism and oversimplification, we can, with good reason and fortitude, concentrate our attention on narrowing the range of disagreement—both as to practical goals and workable (perhaps even acceptable) solutions. Economic stability is not a mirage. I t is attainable, provided we define it as orderly economic adjustment or the avoidance of major eco nomic maladjustments. W hat we really want, and the most we can legitimately expect, is fairly regular overall economic growth (rising real incomes) without having to endure the appalling social wastes and suffering of mass unemployment on one hand or the cruel penalties and social frictions of persistent inflation on the other. A t the same time, of course, we want expanding economic opportunities and indi vidual economic freedom. Stability, then, also demands flexibility and sufficient latitude for those normal and necessary fluctuations to occur in output, prices, incomes, and employment which must take place if the severities of maj or dislocations are to be avoided. In other words, a reasonable policy goal lies in a range of variability among the major economic processes or categories. I f the limits of tolerance are set too tight, or if “stability” is used as a policy justification to shelter particular groups from economic change and adjustment, sta bility of the system in the larger sense makes no sense. All this may seem obvious, but the obvious is most often forgotten. Essentially, flexible stability (or stable flexibility) boils down to the rule of reason—the golden mean between extremes. We cannot set arbitrary targets as to rates of economic growth, or expect these rates to be constant. We cannot overload the economy in response to special pleas during periods of inflation, or grossly mismanage our affairs by uncompromising attitudes during depression. We must adopt a real istic concept of the maximum employment goal of the Employment Act, which some people seem to interpret (in deed, if not in word) as meaning that the Federal Government should guarantee every indi vidual a job doing exactly the kind of work he wants to do (however the consumer may feel about this) and at rates of pay which he indi vidually or collectively demands. We cannot attem pt to force con tinuous full prosperity on every industry, occupation, region, village, and farm at all times without destroying the viability of the system. We cannot drag a crowd of sociological problems, however im portant each may be, under a great tent called “economic stability” without confusing the issues and widening disagreements. In sum, a reason ECONOMIC GROWTH AND STABILITY 327 able stability goal must be viewed, not only as a technical economic problem, but as a problem of social discipline. Turning, now, to the question of means as distinct from the ends themselves, it is again possible to narrow the area of disagreement. During the past two decades a vast literature, theoretical and applied, has mushroomed on questions of stabilization policy. The literature is diverse in policy implications and prescriptions, and much of it deals with narrow technical aspects of the problem, with very limited appli cability to policy questions. Yet even the theoretical discussion pro vides im portant insights, and out of this mass may be distilled some im portant propositions which are highly relevant to policy formula tion and which represent a consensus of the large majority of economists. General stabilization policies may be grouped roughly into two categories: fiscal—having to do with government spending and tax ing; and monetary—concerned with control of the money-supply and credit conditions. Debt management can probably be best char acterized as a marriage of fiscal and monetary policies—a union for which, because of the large size of the Federal debt, no divorce is possible; though the marital relations are not always happy, the partners usually try to put on a brave show of affection in public. The present debate is basically concerned with how these monetary and fiscal policies should be used and in what combination. The his tory of the debate clearly illustrates that the development of economic doctrines certainly follows no orderly growth pattern. In the 1920’s monetary policy held full sway, only to be discarded or scornfully ignored during the 1930’s when fiscal policy became the great hope of stabilization during the prolonged depression. W ith the problems of postwar inflation, monetary policy came back to the fore. I t is now clearly recognized that any one-sided approach to stabilization is ill advised and likely to be ineffectual. Both fiscal and monetary policies must, of necessity, be used in concert. I f they work at cross-purposes, stability depends on the coincidence of for tuitous circumstances, rather than on responsible direction of public affairs. In the past 25 years the role of government in economic life has been greatly expanded. Both in absolute and relative terms, the public sector, as compared to the private sector, has grown so large that for good or ill the fiscal policies of government do have a very significant impact on the economy. Because of the continuing heavy expendi tures for national defense and the wartime legacy of a large Federal debt, what the Federal Government does now and in the future with regard to taxing, spending, and debt management will necessarily loom large in determining an inflationary or deflationary course of events. In short, we have big government, and, therefore, cannot afford fiscal irresponsibility in government. W ith the sobering experience of both a major depression and a great inflation in recent times, coupled with a recognition that big govern ment will continue to play a major economic role whether we like it or not, there is general agreement th a t fiscal policies should be at least conducive to economic stability, rather than instability. To some people, direct manipulation of government spending will always be considered the least appropriate policy device of the whole kit of 328 ECONOMIC GROWTH AND STABILITY stabilizers, but all will agree th at government spending should at least work in the right direction. Few people would knowingly ad vocate that the Federal Government commit fiscal acts which are de liberately unstabilizing in effect—which worsen depressions or add fuel to an incendiary inflationary situation. This rather lengthy introduction, it is hoped, will serve as a point of entry into (rather than a point of departure from) the rocky field of stabilization policy. More narrowly, it is intended to stress the fact th a t government spending policies occupy just one of several tracts in this field, perhaps one of the smaller tracts, but certainly one which cannot be properly cultivated unless the general outline of its boundaries and position is at least roughly delimited. In what follows, an attempt is made to place stabilizing expendi ture policy in perspective. Nothing new has been added, and much is repetition. But merely reviewing the state of the debate and focus ing attention on the main policy implications of current economic doctrine may provide some orientation and clarification for the policy maker. Briefly, this paper attempts to outline the basic issues, summarize the limitations and practical problems, and develop a few general guidelines for government spending policy. This approach, it is hoped, will help to narrow the range of disagreement on how gov ernment spending policy can contribute to the achieving of the im plicit and explicit aims of the Employment Act of 1946—without, it should be added, overriding other equally im portant social consider ations. T h e B a sic P olicy I ssu es The fundamental policy problem in most elemental form is twofold : (1) Should the Government directly manipulate the volume of its expenditures in light of prevailing or expected economic conditions ? (2) I f so, to what extent, according to what criteria, and how should such spending be financed ? The logical'possibilities Out of prolonged debate, several logical possibilities as to fiscal policy and their implications have emerged. In barest outline, they are as follows: Compensatory spending According to this school of thought, the Government should take positive stabilizing action by increasing expenditures during a reces sion to offset, at least partially, declining or deficient private demand for output. Conversely, during boom periods the Government should cut expenditures to reduce inflationary demand pressures. In some formulations the compensatory principle has been linked with the so-called stagnation or mature economy thesis which holds th a t declining investment opportunity in relation to the volume of savings generated in a well-developed economy leads to a chronic deficiency of private investment and therefore to chronic under employment. Under such conditions, it has been argued, government spending (largely financed by budget deficits) is necessary to close ECONOMIC GROWTH AND STABILITY 329 the gap in aggregate demand in order to maintain high levels of income and employment.1 Compensatory spending, in principle, is in no way related to the stagnation thesis. In its usual formulation this policy calls for deficit government spending during the downswing of the business cycle to compensate for falling private investment. During the up swing government spending would be cut back in the face of rising private investment. Budget deficits incurred during recession would De repaid from surpluses piled up during prosperity. Though annual budgets would be unbalanced, no permanent increase in government debt need necessarily occur as the Budget would presumably balance out over the course of the business cycle. Compensatory spending would, it is pointed out, have multiplier effects in both directions. Additional government spending by also raising consumer spending to higher levels would create substantially more new income than the amount of the deficits incurred during recession. Conversely, a reduc tion of government spending would cause or induce a fall in consumer spending and income during a boom and thereby serve as a significant anti-inflationary influence. There are many possible variations of the compensatory principle. Largely, the variants involve qualifications as to kind of spending, methods of financing, and the degree of reli ance to be placed on built-in or flexible features as opposed to ad hoc measures. In fact, other budget policies discussed below have some compensatory features in a broad sense. A stable expenditure budget According to this view, government expenditures should be kept stable, except for certain automatic or built-in variations, such as in relief payments, which take place in response to changes in economic circumstances. Expenditures, it is argued, should be based in each case on an evaluation of social benefits, social costs, and a proper allo cation of resources between public and private uses—determined inde pendently, in the main, of the stability problem. Prim ary reliance for stabilizing action, then, would be placed on either built-in flexibility in the tax structure or adjustable tax planning and on monetary policy. The well known CED stabilizing budget is a variant of this ap proach.2 According to the CED proposal, tax rates would be set so as to provide a balanced budget in normal times of reasonably high levels of economic activity and should be adjusted only to allow for variations in normal governmental expenditures. W ith progressive income taxes as the backbone of the revenue system, tax rates and rev enues would automatically rise during inflationary periods and pro duce a budget surplus. Conversely, during periods of falling income, tax rates and yields would also fall and produce budget deficits. In short, stable government expenditures, coupled with built-in flexi bility, would automatically and promptly keep the fiscal program working in a countercyclical (stabilizing) manner. 1 The stagnation thesis, though an odd grain of truth sprouted in some of the pro argum entation, lias been rather roundly debunked by theoretical an alysis and recent history. 3T axes-and the B udget: A Program for Prosperity in a Free Economy, A Statem ent on 'National P olicy of the Research and Policy Committee of the Committee for Economic Development, 1947. Elaborations on th e basic theme have been embodied in a number o f 'CJEI> publications since the proposal w as first set forth. 330 ECONOMIC GROWTH AND STABILITY Advocates of a more extreme version of the stable expenditures: approach would go further and pursue stabilization by manipulating, tax-rate schedules themselves (either by formula or on an ad hoc basis). This approach has considerable appeal to many economists of diverse political and philosophical inclinations.3 Some even sug gest negative tax rates or rebates in periods of severe depression or rapidly falling income as the correct therapeutic procedure. Annually balanced budgets The view that the budget should always be balanced under all conditions—except, perhaps, in periods of severe national emergency such as a war—has long been a hallmark of sound or orthodox finance. In its most rigid form, this approach requires th at government expendi tures should be promptly reduced or taxes promptly increased to main tain the balance regardless of economic circumstances. Such a pro cedure would contribute to instability. I t would allow inflationary increases in government spending during a boom when incomes and tax receipts are rising (or inflationary tax decreases) and during reces sion call for reduced government spending (or increased taxes) when incomes and tax revenues are falling, and thus worsen the downswing. A more sophisticated version of the annually balanced budget approach is embodied in the so-called balanced budget theorem which theoretically demonstrates how the unstabilizing features of annually balanced budget might be overcome.4 In periods of recession, an increase in Government spending, even though matched by increased taxes, would, under certain highly artificial assumptions, increase aggregate demand and work in a stabilizing direction. Conversely, in boom periods a decrease in Government spending, even though, matched by tax reductions., would decrease aggregate demand. There is nothing mysterious about the “balanced budget theorem.” Assuming that there are no significant indirect repercussions from the changes in fiscal operations on private investment spending and the distribution of income, each dollar of increased Government spend ing adds directly to the Government component of aggregate demand. Each additional dollar added to the tax bill, however, does not reduce private consumer spending by a full dollar, since p art of the taxes are paid from income which would have been saved anyway—that is, not spent on consumption. Looked at another way, the prim ary effect of the increase in Government spending is to increase the total demand and income by the amount of the added spending. In addition, as in the case of compensatory spending, there will be multiplier effects. Consumer spending will rise and create further increments of induced or supplemental income. A t the same time, however, higher taxes work in the opposite direction to reduce disposable income, reduce consumer spending, and absorb it, as it were, or offset the supplemental or induced part of the income created originally by new Government spending. Thus, private spending remains unchanged. Aggregate 3 Cf. A. P. Lerner, An Integrated F u ll Em ployment Policy, P lanning and P olicy for F ull Em ploym ent (P rin ceto n ; Princeton U niversity P ress), 1946, pp. 1 63-220 ; and K. E. Bouldinp. T h e Econom ics of Peace (N ew York : P rentice H a ll), 1946. •. <There is :a la r g e literatu re on the/ m echanics and im plications of the . balanced budget theorem. For a good explanation and references to the literature, see H askll P. Wald, f isc a l; Policy, : M ilitary Preparedness, and P ostw ar Inflation, N ational T ax Journal, II (1 9 4 9 ), pp. 5 1 -6 2 . ....... . . ECONOMIC GROWTH AND STABILITY 331 demand for the economy as a whole is increased by precisely the amount of the increase in the Government-demand component.5 The mechanics are not important for our purposes here. W hat is impor tant is that it is entirely possible to have both higher Government spending and larger tax receipts and yet no change in private spending. The analysis can be reversed and applied to inflation. A reduction in spending matched by an equal reduction in tax revenues reduces the Government component of aggregate demand but leaves private demand unchanged. The uncomfortable alternatives W hat appears to be, at first glance, a wide range of choices open in the matter of stabilizing fiscal policies soon narrows as the alterna tives are examined. A compensatory spending policy involves a considerable expansion of public spending and increased size of the public sector vis-a-vis the private sector during recession. Temporarily, at least, it also in volves deficit finance and increases in Government debt. Such a policy will not be readily espoused by people who believe in limited gov ernment or view government debt with alarm. On the other hand, compensatory spending will also be repudiated quickly during inflation by those who advocate massive increases in the activities and respon sibilities of government. They will find all sorts of reasons why Government expenditures “cannot” be radically reduced and substan tial amounts of debt retired during inflation. As long as some balance of political power is maintained between the two opposing camps, the potential range of compensatory spending may be narrow, indeed, and what compensation is tolerated will likely operate with an infla tionary bias. Politically, it is hard to reduce spending during in flation, especially when costs of Government are rising. Nor is it easy to keep taxes high enough to retire debt. M aintaining stable expenditures and operating flexible fiscal policy from the tax side, likewise, presents uncomfortable problems of choice. Here, the fact that Government spending is supposedly determined on the basis of normal social functions and proper resource allocation, not subject to capricious manipulation, has great appeal. There is no undue expansion or curtailment of the public sector based on ex pediency and crisis, nor is there as likely to be inflationary bias on the spending side. Furthermore, primary reliance is placed on in direct changes in private demand which follow from tax adjustments. There are drawbacks, however. To achieve a given increase or de crease in aggregate demand via tax adjustments alone calls for much larger deficits during recession and much larger surpluses during inflation than would occur with compensatory spending. Also the drastic changes made in rate schedules would have serious repercus sions and create uncertainties that few private persons or public offi cials would readily risk. The balanced budget approach to stabilization also has disturbing implications. Superficially, it has a familiar cloak of apparent finan cial soundness. There would never be a deficit—except when expected 5 For those not fam iliar w ith the uses and lim itation s of an income-expenditure approach to these problems, the balanced budget theorem can be explained in terms of changes in the supply and velocity of money. 332 ECONOMIC GROWTH AND STABILITY tax revenues were overestimated. B ut to balance the budget at high enough levels during a recession to buoy up total demand, calls for a much greater expansion of government spending—increase in the size of the public sector—than does compensatory spending. On the other hand, during inflation the balanced-budget approach m ight re quire such a large reduction in public spending and taxes th a t even the normal social services would have to be curtailed. Furthermore, the tax repercussions would be very large and unsettling indeed. Not only would the prospect of frequent tax-rate manipulations create unstabilizing uncertainties to complicate business decisions, but psy chologically the tax changes required in this case would work in the wrong direction. Rates would have to be raised during recession and lowered during inflation. Such changes could be expected to influence expectations and private investment m a way which would promote instability. On close inspection, there seems to be little appeal in the balanced-budget approach to stabilization except the label. The nature of the alternatives which theoretically (potentially) would achieve the same changes in aggregate demand for the economy as a whole via fiscal policy may be summarized for clarity in the fol lowing tabular m anner: 6 Compensatory spend ing with stable tax rates Stable spending with adjustable tax rates Increase spending... Tax rates unchanged Spending unchanged Decrease tax rates... Increase spending. Increase tax rates. Substantial_______ Little or none______ Greatest. Few........................... Large.............. .......... Many........................ Largest___________ Most. None. Decrease spending... Tax rates unchanged Spending unchanged Increase tax rates___ Decrease spending. Decrease tax rafes. Substantial................ Little or none______ Greatest. Few___ __________ Large_____________ Many_____ _______ Largest...................... Most. None. Balanced budgets I. RECESSION <a) Policy procedure............ <&) Policy implications: Increase in Govern ment spending. Tax repercussions... Size of deficit______ II. BOOM <a) Policy procedure______ <6) Policy implications: Decrease in Govern ment spending. Tax repercussions.-. Size of surplus_____ Clearly, when stripped naked all stabilizing fiscal policies have unsightly deformities which we prefer to keep covered in public and to keep out of polite political discussion. Unfortunately, if they want any stabilizing fiscal policy at all, those who do not like government debt, must, then, be prepared to accept policies which involve the largest fluctuations in government spending and taxes. Those who want minimum manipulation of gov ernment spending must face the largest deficits. Those who honestly advocate compensatory spending must be prepared to fight vigorously for massive reduction of government spending during inflation. In short, in matters of fiscal policy, almost everyone, whatever his prot estations of faith? either unknowing or because he prefers delusion to reality, adopts inconsistent attitudes or demands what is impossible. 6 All of the conclusions embodied in th is table cannot fie derived directly from the pre vious discussion wiiteh sketched only the barest sb rt o f ou tlin e of altern ative fiscal policies. B u t these conclusions do follow from the simple m echanics of ineomfe theory. It^shoiild also be borne in mind th at these conclusions are based on unrealistic sim plifying assump tio n s about the behavior of private demand. The practical value of th is approach is to m ake clear the probable results of certain kinds of governm ent action. ECONOMIC GROWTH AND STABILITY 333 Happily, the theoretical dilemmas themselves exert a centripetal force on groups of differing political and social views and draw them closer together on questions of policy. The range of disagreement is further narrowed by other theoretical and practical limitations to which stabilizing spending policy is subject. To these we now turn. T heoretical and P ractical C om plications The policymaker who is sincerely interested in adapting Federal spending policies to stabilization goals faces a host of theoretical and practical problems. Again in barest outline, some of them are as follows: Oversimplification The theoretical alternatives are based on extremely naive and over simplified models of reality which by themselves provide no basis for policy decisions.7 Simple models of income determination based on a few aggregate variables such as investment, national income, con sumption, etc., mask the complexities and maladjustments which occur in economic life. No amount of manipulation of Federal spending will correct certain kinds of internal maladjustments of a temporary or short-run nature, and such action could conceivably aggravate temporary economic ailments. Moreover, most models reason from ignorance as to the possible interaction of many variables which can not be known or taken into account. Professor Groves has eloquently warned us about the oversimplifications and dangers of aggregative thinking in these w ords: * * * the aggregative point of view becomes frequently guilty of serious omissions. Business and households do not always react along functional lines, and the economic system is not like two quart jars into which one pours economic sub stances until the levels are equal and then all is well.8 The short, naive theoretical models provide useful clues and in sights, but they are mainly useful in telling us what not to do and what to avoid. For example, the balanced budget theorem cannot be used as a policy rule of thumb. Theoretically, it postulates the existence of a single combination of spending and tax rates, which will provide full employment without inflation, and it makes impossible assumptions about the behavior of private spending. Yet it tells us that even a balanced budget may well be inflationary (or defla tionary) and so warns us as to possible effects of budget and tax changes. Forecasting Spending policy, along with all ad hoc stabilization policies, must contend with the well-known problems of economic forecasting. While some improvements in forecasting techniques may be expected, and better statistics more promptly published would be of great as sistance, the problem of guessing future developments will always be 7 For an extended discussion of th is problem, see A. G. Hart, Model Building and Fiscal Policy, American Economic Review, XXXV (1945), pp. 530—548. 8 Harold M. Groves, Financing Government, 4th edition (New York : Henry H o lt), 1954, p. 334. 97735—57----- 23 334 ECONOMIC GROWTH AND STABILITY a frustrating game, especially for the policymakers who have to make the decisions. There is no need to dwell on this point. Suffice it to say, short-run forecasting always will, in the nature of the case, be a slippery proposi tion which will militate against setting policy targets except in the most general terms. The problem is particularly serious for manipu lation of government spending because of the timing problem noted below and the limited scope for automatic or built-in adjustments in the volume of expenditures as economic conditions change. I t should also be remembered th at compensatory variations in spending based on a misplaced certainty about uncertain forecasts can be unstabilizing rather than stabilizing. Timing One of the oft-repeated limitations of manipulating government expenditures is the time element. The budget process is long—too long for prompt adjustment of spending to counteract changes in the private economy as they occur. F or this reason, there are grave doubts that government spending can ever play a major ad hoc role unless things really get out of hand. In our constitutional setup, the Congress cannot abdicate its responsibilities in controlling the public purse. I t cannot, except within narrow limits, give discre tionary authority to the executive branch of Government. Yet, it is well known that once cumulative upward or downward movements get underway, ever larger offsetting increases or decreases in spending would be required—assuming prim ary reliance were to be placed on fiscal policy. While there has been much desultory discussion about a reserve shelf of engineered public works to be promptly started as economic slack appears, there seems little possibility the Federal spending can and will be adjusted promptly and often enough to cushion deflationary developments. On the inflationary side, not only is it difficult to make downward adjustments in government spending to reduce the economic overload (where should the cuts be made?), m the face of rising costs, but the decisionmaking process takes too long to provide prom pt relief. Timelags Closely related to the mechanical problem of policy tim ing and the general oversimplification issue mentioned above are timelags in the economic system itself. Most economic analysis on which stabilizing expenditure policies m ight be based is concerned with different eco nomic positions or effects after certain types of adjustments to changed conditions have taken place. B ut the process of adjustment itself takes time, and there may be long delay before significant economic effects of policy action become apparent. We cannot here go into the various technical problems involved, but one example will suffice to illustrate the complexities. I t may be assumed that if the level of government expenditures is increased by a given amount (say, $2 billion per time period) the stimulative impact on aggregate demand will be greater than the incremental increase in government spending. In this example, total demand would rise by some multiple of $2 billion. This is the well-known “multiplier” process of national income theory. The increase in the volume of government spending creates new demand and new income ECONOMIC GROWTH AND STABILITY 335 in the first instance. In addition, some p art of this new income will be spent by consumers causing demand and income to rise still higher. Successive rounds of consumer spending will follow, each adding successively smaller increments of demand and income until the im pact of the initial increase in government spending has been com pletely absorbed by the system. In the end the new equilibrium level of demand and income is higher than the old level by an amount which is significantly greater than the new government spending. But this process takes time. Because of many factors (such as the savings-spending behavior of the public, income velocity of money, the length of income propagation periods, credit market repercus sions, and many others) the hoped-for stimulative effects may not be felt with any force for 12 to 18 months—by which time economic conditions may have changed. In fact, the changed conditions might well call for policies which would exert the opposite or counteracting effects. Many examples of timelags in the economic processes and events could be cited to reinforce the main observation here: that direct manipulation of government spending aimed at achieving precise (predicted) results presumes a rapidity of economic adjustment which is not possible and cannot be expected. Furthermore, it is presumed that other things will stay equal long enough for the expected results to be obtained. Needless to say, economists are (or should be) rather humble and careful about such presumptions. Minimum sensible changes Another serious problem for the policymaker is to determine how many and what kind of changes are necessary or desirable. This problem is also related to some of those previously raised—oversim plification, timing, and so forth—but merits a separate word or two. Obviously, in an economy with a $400 billion gross national product, a given increase in spending of, say, $5 billion, will have less impact than in an economy with a $100 billion gross national product. Al though we may seem fairly confident at times as to the right direction, no one really knows how much. The problem and some of its impli cations have been summed up neatly by Professor Smithies, though he probablv exaggerates for emphasis, as follows: I f we were properly conscious of the margins of error to which our economic analysis is subject, I doubt whether we would expect that a $5 billion change in any one factor would have an ascertainable effect on total economic activity. This, incidentally, points to a basic dilemma in the use of fiscal policy as an economic stabilizer. To produce decisive results, changes in the budget surplus or deficit of from $10 billion to $20 billion may be needed, and then there is always the danger of overshooting the mark in the direction of infla tion or deflation as the case may be. But budgetary changes of sucli magnitude might disrupt if not debauch the regular budgetary process. In view of the practical limits on changes in the budget, I do not believe that fiscal policy alone can be relied on as an economic stabilizer. But there is no 336 ECONOMIC GROWTH AND STABILITY reason why its influence should not be exerted in the right direction.9 The question of “how much” cannot in practice be separated from the question of “what.” The leverage effects of changed government expenditures will vary with direction or kinds of expenditure. In this connection, it is usual to separate “transfer payments” from “gov ernment purchases” of goods or services. Transfers which merely redistribute income from one group to another may, it is assumed, be mildly stimulative in the short run. To the extent that they tap idle savings and are promptly spent by recipients on consumer goods, some rise in total demand might occur. Redistribution on too large a scale, however, obviously might be self-defeating, aside from the disturbing long-run implications of such a policy. Government purchases of output, on the other hand, are directly income creating. Public works, the object of flattering consideration in the past, now have to compete with national defense and an everincreasmg host of prolific “social priorities” which seem to have tre mendous multiplying powers. One of the “knotty” issues of direct manipulation of government spending lies in the fact that “stability” may well conflict with “social priorities.” The particular increases or decreases in spending which would be most stimulating or tranquilizing for the economy are not necessarily the ones which could easily be adjusted in light of other political and social considerations. Increasing certain kinds of government purchases might have little stimulating effect. F or example, government spending might serve to bid up wages and prices in a particular sector and thereby cause an offsetting reduction in private demand. Conversely, in boom periods a cut in certain types of government spending—research as a possibil ity—might have a negligible effect in reducing total demand. Our wartime experience indicates that leverage effects from d if ferent kinds of expenditures may differ greatly from time to time. To achieve maximum stabilizing action via government spending, there probably should be a good deal of internal shifting in expendi ture programs—not an easy policy to devise or follow. No attempt is made here to assess the conditions which would favor one type of ex penditure as compared to another. The point here is simply to draw attention to one aspect of the problem which receives too little atten tion. The determination o f norms Another problem which must be mentioned, at least in passing, is the determination of norms or criteria for policy formulation. Since criteria for spending policy are being dealt with at some length in other papers in this compendium, only two items will be mentioned. F irst, what constitutes reasonable or realistic “full employment” goals ? This matter was touched on in the introduction. Second, and just as important, is the choice of a general price level goal which is consistent with “normal” (as opposed to stabilizing) changes in government spending and the tax structure. There is real danger that even well-meant policy will place the Government in a never-end ing series of inconsistent positions. The problem is particularly acute „ * •A rthur Sm ithies, The Twin Objectives of Tax Reduction and Reduction of the Budget Deficit, N ational Tax Journal, VXII (March 1955), pp. 30-31. ECONOMIC GROWTH AND STABILITY 337 for a “stabilizing budget” policy which relies heavily on built-in tax flexibility. A tax schedule and tax structure consistent with both high employment and stable prices could be established only by a difficult process of trial and error. When the difficulties of fore casting growth factors are taken into account, the determination of policy criteria becomes an even more complicated problem. Chang ing price levels and rising equilibrium levels of output put the policy maker out on an uncomfortable limb. W hat constitutes an appropri ate high employment goal becomes more uncertain, and margins of error in other estimates likewise become greater. All this really boils down to the not very helpful observation that stabilizing policy would be ever so much more simple if we only had stability. Market discipline The problems of any stabilization policy are intricately related to the operation of the market and the willingness of policymakers to work with the market rather than against it. In periods of inflation, stabilization policy calls for restrictions which always evoke outcries from special interests who will seek Government shelter from market pressures. I t goes without saying that the Government will under mine its own stabilizing procedures if it spends to subsidize indiscrimi nately those groups who feel the restraints. Inflationary price movements are not only the result of economic overload but also are intimately related to the problem of market structures which fiscal policy can do little to solve. When money wage rates are pushed up more rapidly than productivity gains over broad sectors of the economy, costs and prices must rise. Cutting back Government spending may reinforce monetary policy and help to discipline “cost push” inflation, but perhaps only with the pain of some unemployment. Here the basic problem can be solved only by vigorous enforcement of competition in both the product and labor markets.10 In depression, likewise, well-meant attempts to protect particular groups or shotgun tactics may be o f little help and even will impede recovery if they clumsily interfere with automatic compensatory adjustments in the market places.11 The perversity o f State and local finance I t is wyell known that State and local finance may misbehave in a manner similar to sensitive segments of the private economy. In a period of recession, State and local spending governments are faced with declining tax revenues and are forced to reduce total expendi tures, though certain kinds of expenditures, such as relief payments, may rise. During prosperity, on the other hand, State and local ex penditures for construction and public services expand. As a result, Federal stabilizing expenditures are at least partially offset by “wrong way” changes in spending at the State and local level. Moreover, when the Federal Government shows a willingness to increase spend ing on public works, State and local governments have a natural incli 10 Cf. E. Despres, M. Friedman, A. G. H art, P. A. Samuelson, and D. H. W allace, The Problem of Economic Instability, American Economic Review, XL (1950), pp. 505-538 ; also The Mechanics of Inflation (W ashington : Chamber of Commerce of the United S ta tes), 1957 . 11 Cf. Gerhard Colm, The American Economy in 1960 (W ashington: N ational Planning A ssociation), 1952, ch. VI. '338 ECONOMIC GROWTH AND STABILITY nation to allow Federal financing to displace local financing to tlie extent they are allowed to do so. This perversity of State and local finance will probably always be a discouraging, unstabilizing influ ence. While this is a drawback, it provides no argument against Fed eral stabilizing action. I t is just another unavoidable complicating factor—a bad example which the Federal Government should cer tainly not follow. Public attitudes Finally, public attitudes, whether rational or irrational, impose serious limitations on the use of government spending as a stabilizing device. Fiscal orthodoxy which demands annually balanced budgets will always have a strong appeal. There are good reasons for this state of affairs. A balanced budget provides an easily understood rule of cost for transferring resources from the private to the public sector.12 Compensating expenditures and deficits, it is argued, delude the pub lic as to costs and provide a vehicle for ever greater expansion of government. Many, in desperation, see strictly balanced budgets as the only realistic means of keeping government spending in bounds. Who can really say th at this is a naive approach—especially since there are other stabilizing procedures available, such as monetary policy, which may be adequate to cope with the smaller economic fluctuations ? U ntil there is more evidence of fiscal responsibility in government, direct manipulation of government spending to meet changing economic circumstances will he opposed by many as a dan gerous practice. In dire emergency, of course, exigency will dissolve some of the resistance to deficits. In less dire circumstances, resist ance is bound to be stiff. A t least, there is fairly general agreement th at government spending-taxing policies should, at minimum, be neutral—never positively unstabilizing. This represents a real ad vance in the level of economic literacy. I m p l ic a t io n s for S p e n d in g P o l ic y I t is now time to shake some policy implications out of the foregoing estimate of the situation. Some readers will feel th at the outlook for stabilizing expenditure policy is bleak, indeed—that the complexities, problems, and differences of opinion will preclude rational adaptation of Federal spending in ways which will contribute to the attainment of the economic goals of the Employment Act. B ut such a pessimistic conclusion does not follow. The inherent difficulties, the policy dis comforts which arise from extreme positions, and the practical prob lems which must be faced in the real world, all work to narrow the range of disagreement and bring more closely together diverse groups with conflicting views and values. To be sure, there is always danger th at stalemate rather than compromise might occur. Moreover, there will always be disagreements—even major disagreements—but these will arise mainly over questions of degree, the appropriate combina tions of policies, and the economic outlook. There is, nevertheless, good reason to be optimistic—though not complacent—in the hope th at 12 Cf. Jesse Burkhead, The Balanced Budget, Q uarterly Journal of Economics (May 1954), pp. 191-216. ECONOMIC GROWTH AND STABILITY 339 there exists a sufficiently broad area of general agreement on funda mental matters of policy to promote workable and acceptable solu tions. Set forth below are a series of observations based, largely, on the preceding analysis, which may provide some positive guidance for the policymaker. This writer believes, perhaps presumptuously, that they will command fairly widespread assent among most economists and a majority of thoughtful nonspecialists. (1) Government spending is only one element of overall stabiliza tion policy—one which will normally occupy a secondary, or even minor, position, especially in combating minor economic fluctuations. Monetary policy is obviously the first line of defense against insta bility, though it is potentially more effective against inflation than deflation. Monetary policy is general in application, can be promptly applied, and is aimed at stabilizing private spending. Tax policy is at least as important a spending policy as a stabilizer—perhaps more so, because it is more automatic, indirect, and general in application. Also, tax adjustments probably are more widely acceptable than is direct manipulation of government spending. (2) Spending policy, in concert with other fiscal measures, can make a major contribution to stable employment and price levels, as a firm and effective “backstop” to monetary policy. In other words, changes in Federal spending should work in the same counter cyclical direction as credit restraint or credit ease.13 This means that, in recession, some expansion of government expenditures—whether on an ad hoc or an automatic basis—is both necessary and desir able. But this does not mean that, in every lull in the rate of eco nomic growth or in a minor readjustment period, the Federal Gov ernment should rush excitedly into new or expanded spending schemes to encumber the economy with unneeded help. During inflation, on the other hand, the Federal Government should adopt a rigorous prosaving attitude toward spending which makes every Government activity and project “fight for every dol lar”—to use one of the late Professor Schumpeter’s vigorous phrases. This does not mean that inflation can be used as an excuse to cut all types of spending. Basic governmental services are either necessary and appropriate or they are not. Each should be decided on indi vidual merits. But it does mean that, where cuts cannot properly be made or expansion of spending is deemed absolutely necessary, taxes must be raised high enough to cover all spending and provide for a substantial budget surplus, as well. Stabilizing expenditure policy cannot be a 1-way street and still merit public support. I f increased spending is necessary during recession, reduced spending is likewise necessary during inflation. Clearly, monetary policy cannot be expected to serve as an economic policeman, to guard against disorder during recession and to arrest inflationary excesses during boom, if the superintendent of police—the Congress—in fiscal matters fails to give support in hard times and tears up most of the arrest tickets during inflation. 33 Because monetary policy can be reversed more quickly th an spending policy, and be cause changes m ust be made occasionally in the “norm al” level of Government spending for other reasons, there will be times when m onetary and spending policies will, and quite properly should, work in opposite directions. 340 ECONOMIC GROWTH AND STABILITY (3) Adequate budgetary procedures and control are essential in gredients of stabilizing spending policy. Long-run spending policies as to government services, programs, and responsibilities should be decided on the basis of what constitutes the proper agenda of Govern ment, “social benefit-social cost” considerations and optimum resource allocation, without direct reference to the problem of economic in stability. Rational adjustment in government spending, however, presumes some initial ordering of priorities and perhaps subsequent reordering as economic conditions change and the questions of eco nomic impact must be taken into account. Budget preparation and procedure should make it possible for the Congress to evaluate spend ing proposals—to select the least im portant and pressing programs which can be eliminated during inflation or postponed to be under taken during periods of economic slack. Likewise, better budgetary procedures would help eliminate make-work spending schemes during depression which are not consistent with long-run spending plans. (4) Serious consideration should be given to ways and means by which a greater degree of automatic flexibility could be built into Government spending programs, without, of course, weakening con gressional control of the public purse. This is a difficult and touchy issue which has often been discussed and dropped. But the fact re mains that until some way is found to overcome the problem of selec tion and timing, the potential contribution of government spending to stabilization policy will be seriously constricted. There is already some built-in flexibility in certain kinds of government expenditures, mainly in transfer payments for relief and in farm subsidies. In principle, it should be possible to build substantial countercyclical flexibility into some other types of expenditures, especially in longrange, social-investment programs which can be adjusted to changing economic conditions without creating serious hardship. The present long-term, Federal highway program is a case in point. I t has been suggested that this program, as it is now set up, may have perverse effects on economic stability. Basically, under the trust-fund arrangement, spending from the fund is geared to tax receipts flowing into the fund. I f tax revenues fall, construction would be delayed. Although there is some discretionary authority on the p art of the Sec retary of Commerce to release additional money when the tru st fund has a surplus, regular congressional action would be required to make temporary appropriations (loans) from the general fund to the trust fund before spending could exceed current receipts and surplus. This w riter is not familiar enough with the details of the Federal highway program to know how it will work out in practice, or to evaluate the suggestions th at the program m ight work in an unsta bilizing manner. This program has merely been singled out as an ex ample of long-range social-investment spending which has already been undertaken on its own merits and which could be so rigged as to make a positive, rather than a negative, contribution to economic stability. State and local governments simply cannot regularize their expen ditures, but in joint Federal-State ventures they could be helped to do so. Furthermore, there is no reason why long-range, Federal spending on public works should follow the perverse pattern of State and local finance. Surely, formulas could be devised, consistent with adequate budget control, to make Federal spending a more sensitive ECONOMIC GROWTH AND STABILITY 341 and automatic instrument of stabilization policy. I t will not be easy to get agreement on how this can be done, but solutions in this direc tion should be patiently and persistently pursued. (5) Annually balanced budgets, or overly balanced budgets to pro vide for secular reduction of debt, are a necessary policy rule in nor mally prosperous times. In times of economic stress, much beyond a mild recession, the Federal budget simply cannot be balanced without making recession more severe. Tax revenues will drop much more rapidly than expenditures can be reduced. Moreover, even if it were politically possible to cut spending at such times, few people would suggest that public spending should be cut simply because private de mand was falling, or that the Government add to downward pressures. A recognition of the fact that under deflationary conditions the rigid insistence on an annual balance in the Federal budget would make the Government a promoter of depression throws out of court schemes which call for debt retirement according to a fixed annual schedule or formula. While systematic secular retirement of debt, when economic conditions are favorable, is a desirable policy goal, plans for debt retirement must be flexible enough to allow the Federal Government to discharge its minimum responsibilities under the Em ployment Act. B ut a recognition of the fact that an annually balanced budget is neither possible nor desirable under adverse economic conditions also makes it essential for the administration and the Congress to take forthright steps to eliminate the inflationary bias which seems, inevi tably, to creep into fiscal policy. I t is, of course, politically difficult to reduce spending, to keep taxes high, and to retire substantial amounts of debt during periods of inflation, but public support for stabilizing fiscal policy can be greatly widened and strengthened if there is also public confidence that fiscal discipline can really be expected even in election jears. “Social priorities” and “emergencies” cannot be used as meaningless catch phrases to justify any and every scheme for ex panding governmental programs without reference to costs—costs both in terms of the resources required and the costs of inflation. Eco nomic overload by government contributes to inflation and a course of economic events which is diametrically opposed to the goals of the Employment Act. Furthermore, many well-meaning and thoughtful people distrust and oppose compensatory finance, built in or pushed in, because they see in it the mechanism for ever expand ing the scope of government under the appealing guise of “stabiliza tion policy” whereby government spending is expanded in depression and expanded some more during inflation. I f Federal spending is to make its important, albeit limited, contribution to greater stability, the Congress must demonstrate that the goals of the Employment Act, including the implicit goal of fairly stable price levels, are more than a pious declaration of faith. (6) Because of the many problems, theoretical and applied, involved in the direct manipulation of Government spending, the financing of expenditures—tax policy—must play a major policy role. Built-in flexibility in the form of income taxes makes a healthy contribution to stability, especially in dealing with minor fluctuations. In addition, tax rate adjustments should be used judiciously to augment and rein force spending policy. By working from both sides of the fiscal equa 342 ECONOMIC GROWTH AND STABILITY tion, some of the serious dilemmas of extreme positions can be allevi ated, although never entirely eliminated. Frequent changes in the tax code are not desirable because of administrative difficulties and the unsettling repercussions they have on the private economy. Business decisions are difficult at best without creating additional uncertainties. F or this reason, it behooves us to develop a tax structure which per mits tax rate adjustments to be made with minimum adverse effects on business decisions and private demand. Tax problems are beyond the scope of this paper, but they cannot be divorced from the expendi tures problem in rational policy formulation. P o s t s c r ip t No attempt has been made in this paper to be comprehensive or to innovate. An attempt has been made throughout to maintain a policy, rather than a technical, orientation. As was stated at the outset, the purpose of the paper has been to discuss the issues and problems with a view to staking out a sizable area of agreement on policy goals and procedures. The summary of policy considerations immediately above is far from being either exhaustive or precise. B ut it may provide some guidance on difficult and important questions. Fortunately economics has no political affiliation—though the public is no doubt often perplexed on this score. No amount of economic analysis, however thorough and relevant, can provide ready policy prescriptions or relieve us from the inescapable, and often uncom fortable, necessity of choosing among various alternatives. Economics is a dismal science because it tells us that we cannot have our cake and eat it too. I t becomes less dismal when we realize th a t economic analysis may equip us to choose more intelligently. I f this paper has helped in a small way to clarify certain choices and their im pli cations with respect to Government spending policy, its addition to the vast existing supply of printed pages dealing with the same subject may be justified.
@FedFRASER