The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
FEDERAL RESERVE B A NK O F ST. L O U IS FEBRUARY 1972 The Economy in 1972 ............................ 2 Operations of the Federal Reserve Bank of St. Louis— 1971 ................... 13 Projecting With the St. Louis Model: A Progress Report ............................. 20 Vol. 5 4 , No. 2 The Economy in 1972 by ROGER W. SPENCER 7^_LTHOUGH most projections of the pace of economic activity this year are quite optimistic, un certainty continues to cloud the 1972 economic hori zon. With unemployment still high by historical standards, many people remain concerned about their ability either to obtain a job to their liking or to retain their present one. Inflation and potential inflationary pressures are not yet subdued. The effectiveness of the price and wage control program is still debated by those who regard it as a necessary, long-term supple ment to orthodox monetary-fiscal policies, and by those who see the program as a disturbing encroachment on individual freedom. Nineteen seventy-two could also be a year marking the most sweeping changes in the international monetary payments mechanism since the 1944 Bretton Woods Conference. Finally, all these economic developments take on added significance in light of the coming presidential election. Many of these issues evolved over a long period and will be difficult to resolve in a single year. To assess the course of economic activity over the next several months, this article discusses first the principal influences on the course of economic activity and secondly, the economic outlook. Factors Influencing The Economic Outlook The chief factors to be considered in projecting the economic outlook for 1972 are given in Figure I. The three major categories are cyclical forces, structural changes and policy actions. At any time, the economy has a certain “momentum” of its own. The direction and magnitude of the mo mentum reflect the particular stage of the business Page 2 cycle. The momentum of cyclical forces can be offset or augmented by structural changes or stabilization policy actions. Structural changes include developments in the world trade picture which influence the U.S. balanceof-payments position, random events such as wars, strikes and weather shifts, as well as “other” factors. These other factors include: (1) changes in prefer ence, such as an increased preference for saving rela tive to consumption, or liquidity relative to goods and services; (2 ) changes in the nature or rate of in crease of technological advance; (3 ) changes in the quantity and quality of available resources, such as a shift toward more availability of inexperienced labor resources relative to experienced labor resources; and (4 ) changes in the institutional and legal framework, such as the initiation of legally authorized price and wage controls in a peace-time economy. Policy actions are the orthodox measures employed by stabilization authorities to guide the course of eco nomic activity. Monetary and fiscal actions taken prior to and during 1972 will have a substantial bearing on economic developments during the current year. Cyclical Influences Nineteen seventy-two begins the second year of re covery from the moderate recession which ended in November 1970. Much has been made of the fact that the first year of the current recovery period was not as strong as the first year of previous recoveries, but insufficient attention has been given to the relation between the moderate recession and the moderate recovery. To the extent that forces are set in motion F E D E R A L R E S E R V E B A N K OF ST. LO UIS which influence the scope of business cycles independ ent of policy actions or other outside forces, a strong recovery in 1971 should not have been anticipated.1 Indeed, many observers accurately predicted a mod erate recovery in 1971 based, in some degree, on such reasoning. iW esley C. Mitchell, a pioneer in business cycle analysis, wrote on the automatic forces influencing the up and down phases of the cycle; The various processes just described combine reduc tions in both prime costs and fixed charges with an ex pansion in the physical volume of business. In this fashion depression ultimately brings about revival. For of course these changes increase prospective profits, and in the money economy prospective profits are the great incentive to activity. . . . F E B R U A R Y 1972 Nevertheless, the weakness of the recovery last year has contributed to the view that there has been a change in the way the economy works, that the economic principles which held in times past no longer apply. There is no question that the economy is con stantly changing and the policies used to influence the course of its movement must be reasonably flexi ble. However, a simple comparison of the moderate recovery in 1971 with vigorous first year recoveries in the past may overstate the degree of structural change, especially with regard to production and employment. In fine, this business situation is that described in the first section of Chapter 1 — the situation out of which a revival of activity presently develops. Having thus come round again to its point of departure, after tracing the processes of cumulative change by which prosperity breeds crisis, crisis evolves into depression, and depression paves the way for a return of prosperity, the present theory of business cycles has reached its appointed end. [Wesley Clair Mitchell, Business C ycles an d T heir Causes (B erke ley and Los Angeles: University of California Press, 1 9 5 0 ), pp. 146-47.] This volume is a reprint of Part III of M itchell’s Business C ycles, originally published in 1913 by the University of California Press. Table I presents changes in several important eco nomic variables for each business downswing, for each recovery and for each entire recession-early recovery period of the past two decades (see also the ac companying chart). The “Total Period” data reflect both the depth of the recession and the strength of the early portion of the ensuing recovery. On this basis, the recent economic experience docs not appear as strikingly different from other periods as a com parison of recessions alone or early recovery periods alone would suggest.2 There is some likelihood that, other things equal, the more severe is the downturn due to such factors as described by Mitchell, the stronger will be the upturn. For example, if the downturn is of sufficient scope that even very efficient resources become unemployed, the efficient resources can be utilized to produce a strong upturn after the crisis point is passed. Outside forces, however, such as strikes and policy actions, can effectively alter this relation. 2Most recessions are often termed “inventory” recessions since much of the adjustment in total spending is reflected in the rate of inventory accumulation. The 1969-70 recession differed in the respect that inventories did not fall as they did in previous recessions. Accordingly, it should not have been expected that inventories would have increased significantly in the early recovery, and they did not. The change in real ( price-deflated) inventories averaged $2.3 billion in the Page 3 F E D E R A L R E S E R V E B A N K OF ST. LO UIS F E B R U A R Y 1972 T a b le I Recent Recession and Early Recovery Periods ( A n n u a l Rates o f C h a n g e ) In d u stria l Production Real Product Com parison of Four Economic Indicators Recession— Recovery T rough s = 100 P a yro ll Em p lo ym e nt GNP Deflator R h | pr0(jl|C, 110 105 R E C E S S IO N 111/53-111/54 111/57-11/58 — 7 .6 % — 1 4 .5 T ro u gh s = 100 — 1 .6 % — 3 .2 % 1 .4 % -4 .6 — 5 .0 110 3RD C TR'54 2 N D )TR 58 1ST Q TR '67 4TH C TR.70 . 2.3 li/ 6 0 -1 / 6 1 — 7 .3 — 1.9 — 2 .4 1.6 IV / 6 9 - IV / 7 0 — 6 .3 — 1.3 — 1.0 5 .7 105 1969-71 100 100 RECO VERY 111/54-111/55 1 4 .7 8.6 4 .7 1.6 11/ 5 8 -1 / 5 9 2 0 .2 8 .9 4 .5 1.9 1/61 - IV / 61 15.1 8.1 2.8 1.0 3 .4 5 .0 1.3 3.3 0 .7 A v e r a g e of 1 95 3 -5 5 , 1957 -59, a n d 1 960-62 1.5 IV / 7 0 - IV / 7 1 1 95 1 . Industrial 120 95 > roduction 120 3 T A L P E R IO D 111/53-111/55 2.9 3 .4 111/5 7 -1 / 5 9 1.4 2 .0 -0 .4 2.1 II/ 6 0 - IV / 6 1 3 .3 3 .0 0.1 1.3 IV / 6 9 - IV / 7 1 — 1.6 1.8 0.1 4 .5 115 ✓ / Notes: The early recovery period is arbitrarily dated the same length as its accompanying recession ; thus, the first and last recovery periods (1954-55 and 1970-71) are one year in duration and the other two are three quarters. The periods of recession used in the article have been defined by the National Bureau of Economic Research. Because of the automobile workers strike in late 1970, the depth of the most recent recession and strength of the recovery are somewhat overstated. A 1967=100 base was used for the industrial pro duction calculations, except for the first recession-recovery period, in which a 1957— 59 = 100 base was used. 1969-70 recession, compared with an average decline for the three previous recessions of $2.9 billion. The change in real inventories averaged $2 billion in the 1970-71 recovery, com pared with a $3.8 billion average for the three previous recoveries. Some analysts have suggested the current “low” inven tory/sales ratio indicates large increases in inventories in the near future. The inventory/sales ratio is currently “low” only in relation to the past four years. It is not far from the average in relation to past early recovery periods. The average of the inventory/sales ratio in the early recovery period (through November 1971) following the 1969-70 recession was 1.56 compared with 1.54, 1.56 and 1.50 for the recoveries immediately following the recessions of 1960-61, 1957-58 and 1953-54, respectively. Page 4 115 ✓ / ✓ 110 * A v e r a g e of 195*3-55, 1 95 7 -5 9 , / a n d 19< 0-6 2 ^ \ 105 110 ✓ X \ \ / \ N X \ V 1 95 \ / \ / 1 1 / 105 / \ *** ^ j 100 Production and Em ploym ent - Industrial production rose about 20 percent in the first year following the 1957-58 recession, compared with a 3.4 percent in crease in the year following the 1969-70 recession. H o w ev er, in d u stria l production fell at a 14.5 percent annual rate during the 1957-58 recession, compared with a 6.3 percent decline in the 1969-70 recession. Table I indicates that there has been a strong relation between the severity of a recession and the strength of the immediate recovery. In general, the most severe recession of the four compared was that which oc curred in 1957-58, and the strongest immediate re covery followed the 1957-58 recession. The mildest recession was the most recent one (1969-70), and the mildest recovery was in the 1970-71 period. ✓ 1969-71 100 1 1 1 95 P a yro ll Employm ent 105 ^ 100 105 1 95 3-55, 195 7 -5 9 , an d 1 9 6 0 -6 2 *4 ^ ^ ^ ___________ ^ ^ ___ 100 1969-71 . l 95 95 GNP P rice Deflator 105 105 100 100 A v e r a g e of 1 95 3-55, 1 95 7 -5 9 , a n d 1 960-62 95 95 90 - ----------------------------4 - 3 - 2 - 1 0 1 2 3 90 4 QUARTERS TO A N D FR O M TROUG H Note: A 1967=100 b a s e w a s u se d for the in d u stria l production calcu latio ns, except for the first re c e ssio n -re c o v e ry period , in w hich a 1 9 5 7 -5 9 = 1 0 0 b a s e w a s use d . Sources: U.S. D e p a rtm e n t of Labor, B o a rd of G o v e r n o r s of the F e d e ra l R e se rv e System , a n d U.S. D e p a rtm e nt of C o m m erce Latest d a ta plotted: 4th qu a rte r 1971 F E B R U A R Y 1972 F E D E R A L R E S E R V E B A N K OF ST. LO U IS Growth in real product or industrial production has not been as strong in the 1969-71 recession-recovery period as in earlier comparable periods, but differ ences (particularly with the 1957-59 period) are not as great as one might expect. Although industrial pro duction has not yet reached its 1969 peak, real product gains over the latest recession-recovery period are only slightly less than in the earlier comparable periods. Payroll employment changes over the several pe riods are quite comparable. Payroll employment rose at a 0.1 percent annual rate in the 1969-71 period, compared with a 0.1 percent rate rise in the 1960-61 period, a 0.7 percent rate rise in the 1953-55 period, and a 0.4 percent rate decline in the 1957-59 period. Prices —One of the most substantial differences among the indicators and periods given in Table I is the movement of prices in the most recent period. The implicit GNP deflator rose in the 1969-71 period at a 4.5 percent annual rate, more than twice as rapid as in any earlier comparable recession-recovery period. On its face, such a difference would appear to reflect substantial structural changes in the economy. An alternative, or supplementary, explanation for the difficulty in restraining price advances in the most recent recession-recovery period is framed in terms of price anticipations. The longer that prices are per mitted to rise unchecked, the more expectations of rising prices are incorporated into current prices through higher wages, rents, and other contractural obligations. Anticipations of higher prices built up over a period of years can effectively impede the efforts of stabilization authorities to halt inflation with out any significant changes in economic structure (for example, more monopoly power on the part of unions or businesses) having occurred. Prices rose for 35 quarters from the trough of the 1961 recession to the peak of the recovery ending in 1969. The next longest period from trough to peak (in the past two decades) in which prices were per mitted to rise was the 15 quarter period preceding the 1953-54 recession. Given the length of the period over which prices rose and the fact that price in creases accelerated in each succeeding year after 1962 (especially in the latter half of the 1960s), it should not have been surprising that the moderate stabiliza tion measures adopted in 1969 to stem inflation were not immediately successful in reversing the trend.3 :!The model of the Federal Reserve Bank of St. Louis, pub lished in the April 1970 issue of this R eview , indicated at that time that following a course of 6 percent money growth ( and high-employment expenditures as estimated by this B an k) from the fourth quarter of 1969 to the fourth quarter Structural Influences There are sufficient differences in the economic indicators over the recession-recovery periods to sug gest that cyclical influences alone cannot explain the variations. The structural changes which have oc curred over the years have affected, in particular, the short-run unemployment-prices relation. The unemployment rate rise from 3.6 percent at the beginning of the recession to 6 percent at the end of the early recovery over the latest recession-recovery period is the most severe of the past four such periods, despite the fact that payroll employment increased more rapidly in the most recent period than in 1957-59, as rapidly as in 1960-61, and only slightly less rapidly than in 1953-55.4 Part of the explanation for this ap parent paradox is that (1) the labor force has been increasing more rapidly in the past recession-recovery period than in most earlier ones, and (2) the compo sition of the labor force has been shifting such that proportionately more of the labor force is composed of individuals with a historically high average rate of unemployment. Women and teen-agers, for example, comprise relatively more of the labor force than in earlier recession-recovery periods. The change in the composition of the labor force over the past decade represents a structural shift in the economy. It is a shift which has worsened the tenuous, short-term relation between unemployment and prices through the aggravation of unemployment. of 1971 would find prices still rising at a rapid rate at the end of 1971. By late 1971, total spending would be increasing at an eight percent rate with such monetary actions [6 percent annual rate of increase of money]. The rate of price increase would fall somewhat, however, because of past restrictive monetary actions. But the gain in price performance would be small, because by late 1971 prices would still be increasing at a four percent rate [Leonall C. Andersen and Keith M. Carl son, “A Monetarist Model for Economic Stabilization,” this R ev iew (April 1 9 7 0 ), p. 20], A simulation of the model over the IV /1969-II/1971 period ( before price and wage controls) with actual money and high-employment expenditures and coefficients estimated through IV / 1969, projected a 4.9 percent six-quarter average price increase, compared with an actual average of 5.4 per cent. See Keith M. Carlson, “Projecting W ith the St. Louis Model: A Progress Report,” this issue of the R eview , pp. 20-27, for other comments on the model’s predictive performance. 4T he rise in the unemployment rate from 3.6 percent in IV /1969 to 6 percent in IV /1971 reflects the persistence of relatively high levels of unemployment during the recent re covery. LTnemployment rose from 2.7 percent of the labor force in III/ 1953 to 4.1 percent in III/ 1955, an increase of 1.4 percentage points. Unemployment increased 1.6 percent age points from 4.2 percent to 5.8 percent in the 1957-59 period and one percentage point from 5.2 to 6.2 percent in the 1960-61 recession-recovery period. Page 5 F E D E R A L R E S E R V E B A N K OF ST. LO UIS Unem ploym ent Rate Percent 8 1950 F E B R U A R Y 1972 Percent 8 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 Sh o o e d a re a s reoresent p e rio d s of business recessions a s d efine d b y the N a t io n a l Bureau of Economic Research. * D a ta re vise d from J a n u a ry 1970 Latest d a ta plotted: J a n u a ry Another structural change which may have affected unemployment is the fact that U.S. exports relative to imports have been slowing in recent years. Net U.S. exports slowed each year (except 1970) from $8.5 billion in 1964 to $0.7 billion in 1971. Employ ment and output in a number of U.S. industries are probably less than they would be otherwise if it were not for the deterioration of the U.S. competitive posi tion over the past several years. Over a long period, U.S. resources may shift from industries with declining export demand to other industries, but during the period of adjustment/ employment is probably af fected adversely. Quite likely, structural changes have also hampered efforts to slow price increases. The shifting composition of final output from the production of goods to the production of services has probably tended to rein force price pressures. Because of the nature of many services rendered, such as those of barbers, lawyers, and government workers, productivity gains are dif ficult to achieve (or measure). The production of goods such as automobiles or appliances can be more easily enhanced by technological advances and/or mass assemblage techniques. Lower productivity, other things equal, results in lower potential output, a more severe total supply constraint relative to total de mand, and consequently, higher prices.5 The minimum wage, which puts a floor under wage rates, and also tends to increase unemployment by making inexperienced workers ineligible for many jobs, has been extended to cover more workers in re cent years than in the 1950s. Although the minimum wage was designed to benefit the worker, it is possible 5See Roger W. Spencer, “Population, T he Labor Force, and Potential Output: Implications for the St. Louis Model,” this R ev iew (February 1 9 7 1 ), pp. 15-23. Page 6 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 Source: U.S. D e partm ent of L a o o r that it has affected adversely both unemployment and prices. This structural shift is one endorsed by law. Because structural changes probably have accounted for some worsening of the prices-unemployment rela tion, structural measures could play a strong role in improving the relation. Job training, information and relocation subsidies, and the elimination of legal and institutional barriers to jobs (such as the minimum wage) could bolster orthodox monetary and fiscal stabilization techniques to achieve the goals of a low rate of unemployment and price stability. Stabilization Policy Actions The impact of monetary and fiscal policy actions economic activity is, like cyclical and structural influences, not easy to identify or measure. A rough approximation of aggregate monetary and fiscal policy actions prior to and during recent recession-recovery periods is given in Table II. Such actions normally influence economic activity with some lag, and a period of three quarters before each peak was arbi trarily selected as the point from which actions affect ing activity over the course of the recession-early recovery periods should be dated.8 Money, defined to include demand deposits and currency in the hands of the public, is used to represent monetary policy actions and high-employment expenditures is selected as the fiscal policy variable. Both are deflated by the implicit price deflator to account for varying price trends over the past twenty years.7 on °See Andersen and Carlson, “A Monetarist Model,” for some evidence on the relevant length of policy action lags. 'Another method of adjusting stabilization policy actions for changes in trend over a long period is given in Leonall C. Andersen, “A Monetarist View of Demand M anagement: The United States Experience,” this R ev iew (Septem ber 1 9 7 1 ), p. 9. Andersen adjusts the money supply by comparing its growth relative to various trends over the 1952-71 period. F E B R U A R Y 1972 F E D E R A L R E S E R V E B A N K OF ST. LO U IS T able II Monetary and Fiscal Trends Prior to and During Recent Recession-Recovery Periods (A n n u a l Rates o f C h a n g e ) M oney/ Prices Period I V / 1 9 5 2 -111/1955 ......................... 1 .0 8 % E x p e n d itu re s/ Prices — 3 .7 % IV / 1 9 5 6 - 1 / 1 9 5 9 ......................... — 0 .8 1 6 .4 III/ 1 9 5 9 - IV / 1 9 6 1 ......................... — 0 .9 5 4 .3 1/19 6 9 - IV / 1 971 ......................... 2.3 0.21 Rapid growth of the money stock, relative to prices, is presumed to have a short-run stimulative effect on the economy, as is rapid growth of high-employment expenditures relative to prices. For example, there would probably be general agreement that monetary actions were stimulative in the IV/1967-1V/1968 pe riod when money/prices rose 3.2 percent and restrictive in the IV/1968-1V/1969 period when money/prices declined 1.3 percent.8 Table II suggests that monetary actions were rela tively stimulative in the 1952-55 period, but fiscal ac tions were not (unwinding of the Korean War). Fiscal actions were relatively stimulative in the 1956-59 and 1959-61 periods (compared to the 1952-55 and 196971 periods), but monetary actions were not. Neither monetary nor fiscal actions were overly expansive in the 1969-71 period. By these crude indexes, the slightly stimulative monetary actions in 1969-71 relative to 1956-59 or 1959-61, would tend to exert more inflationary pres sure than in the two earlier periods, and also tend to maintain the unemployment rate at a relatively lower level than in the two earlier periods. This most recent monetary trend may represent a partial explanation for the lack of success in slowing price rises in recent years relative to the two earlier periods, and for the success attained in holding the aggregate unemploy ment rate lower than in the two earlier periods (the unemployment rate reached a peak of 6 percent in the latest recession-recovery period, compared with 7.4 percent and 7 percent for the 1957-59 and 1960-61 periods, respectively). Two other factors — the shifts in composition of the labor force mentioned above and the sizable defense cutbacks in recent years (reflected in the 2.3 percent figure for expenditures/prices in the table) —have contributed to the high unemployment rates for in 8It should be noted that monetary and fiscal authorities can not control money/prices and expenditures/prices over any brief period; they may, however, control money and expenditures. experienced workers across the country and defenseoriented workers in particular geographic areas of the United States.9 The Economic Outlook The current economic outlook is based on recent and projected cyclical, structural and policy influ ences. These factors affect economic activity in 1972 and beyond. Current Cyclical and Structural Considerations The U.S. economy begins the year 1972 at a position on the business cycle not substantially dif ferent from that a year ago in terms of actual “mo mentum” relative to potential. Although real product rose at a 6.1 percent rate from the third to the fourth quarter of 1971, after increasing at a 3 percent rate from the first to the third quarter, output advances have not been sufficient to reduce unemployment. The rate of utilization of both labor and capital re sources has changed little in the past year. Actual output is approximately 93 percent of potential out put, about the same as a year ago. Thus the “mo mentum” of the economy moving into 1972 gives little reason by itself to alter economic projections from the actual developments of 1971. °“The total employment attributable to military expenditures, including military forces and government civilian personnel, dropped from a peak of 7.8 million in 1968 to about 6.1 million in 1971, a loss of 1.7 million jobs over the three-year period, with more than 900,000 of these lost during 1970-71.” Richard P. Oliver, “Employment Effects of Reduced Defensed Spending,” M onthly L a b o r R ev iew (D ecem ber 1 9 7 1 ), p. 4. F E D E R A L R E S E R V E B A N K OF ST. LO U IS F E B R U A R Y 1972 Most structural changes occur slowly over time, but there are some important developments changing the 1972 outlook from that of 1971. Little is currently known as to whether preferences for liquidity and saving relative to the purchase of goods and services have changed much from last year, but it is unlikely that the long-term trend of preference toward services from goods will be much altered. Moreover, there is little basis on which to project changes in economic activity in 1972 over 1971 because of differing re source trends, weather changes, war developments, technological advancements, or strike patterns. There are at least two structural changes which will likely exert some influence on the economy in 1972. First, the anticipated improvement in demand for U.S. exports relative to imports should have an expansionary effect on total spending, output, and employment.10 Second, price and wage controls should contribute to a slowing in the rate of increase of prices in 1972, given moderate stabilization actions. Price increases leveled off in 1970 and 1971, and may have been decelerating slightly when the Presi dent called for a wage-price freeze last August 15. Since the imposition of the freeze and the second phase of the Administration’s program, measured prices have slowed. Wholesale prices of industrial commodities increased at a 0.5 percent rate from August to December, after rising at a 4.6 percent rate in the preceding eight months. Consumer price in creases slowed to a 2.5 percent rate from August to D ecem b er, co m p ared w ith a 3 .8 p e rce n t ra te from D ece m b er 1970 to A ugust. Problems of administration and equity will likely intensify the longer the program remains in existence, but in the latter months of 1971, at least, wage and price controls seemed to have had the desired effect. The potential for eventual success of the program is enhanced by: (1) the fact that the controls have not been accompanied by excessive monetary stimulus; (2) the fact that there is currently substantial eco nomic slack; (3 ) the possibility that controls may have helped stem anticipations of higher prices; and 10“The expected employment gains for the United States will probably be spread over about two years, too, according to Peter G. Peterson, the W hite House International Economic Policy Chief. E ach $1 billion of payments turnabout will create 60,000 to 80,000 jobs, he estimates. Thus, attaining the Connally goal of a $9 billion swing for the better could mean more than 700,000 additional jobs by late 1973; that in itself would be enough to lower the unemployment rate to about 5.2 percent from November’s 6 percent,” Richard F. Janssen, “The New Dollar: Devaluation of 8.57 Percent Likely to Create Jobs, Help Nixon Summitry,” T h e W all Street Journ al, Decem ber 20, 1971, p. 25. These exportrelated employment gains anticipated by the administration seem overly optimistic. Page 8 (4) the fact that the wage negotiation calendar for 1972 is relatively light. Recent Monetary and Fiscal Policy Actions The course of economic activity in 1972 depends, in part, on both monetary and fiscal policy actions. There is a carry-over effect from actions taken last year, and there will be some effect from actions taken during 1972. Monetary Actions —The rate of growth of the money stock fluctuated more widely in 1971 than usual. After rising 5.4 percent from December 1969 to December 1970, the money stock accelerated to a 10.3 percent annual rate of growth in the first seven months of 1971, and then slowed markedly to very little growth during the last five months of the year and into January 1972. These violent swings in money movements were accompanied by roughly similar changes in interest rates. The three-month Treasury bill rate, for exam ple, rose from a low of 3.3 percent in March 1971 to a peak of 5.5 percent in July, and then fell to a little over 3 percent in January 1972. Falling interest rates in the latter half of 1971 have been viewed by many analysts as a spur to economic activity in 1972, and indeed a lower cost of capital is often one of the ways by which monetary changes influence spending. F E D E R A L R E S E R V E B A N K OF ST. LO U I S F E B R U A R Y 1972 Fiscal M easures M o n e y Stock (+)Surplus; (-)Deficit Q u arterly Totals at A n n u a l Rates Se ason ally Adjusted Billions of Dollo ions of Dollar! H ig h - E m p lo y m e n t P erc e nta ge s o re a n n u a l rates of chan ge for p e rio d s indicated. Latest d a ta plotted: January estimated However, because there are channels other than interest rates through which monetary growth affects economic activity, the recent slower growth in the money stock may have a restrictive effect on spending (assuming a relatively stable demand for money). When the money stock expands rapidly relative to de mand, its value falls, and people exchange money for goods and services at a rapid rate. Conversely, when growth of the money stock slows, relative to demand, its higher value makes people more reluctant to ex change their cash for goods and services. Persistent and pronounced swings in monetary growth rates have consistently led similar swings in spending. Thus, the slower monetary expansion during late 1971 may retard economic activity during the first part of 1972 from what it would otherwise have been. Activity later in the year will be influenced by mone tary expansion during the early part of 1972. F iscal Actions — Federal expenditures affect eco nomic activity in two ways. First, Federal Govern ment outlays, whether financed by taxes or borrowed from the public, have an important short-run stimula tive effect on total spending. Over longer periods of time, such expenditures tend to displace private pur chases of goods and services. Second, increased Fed eral Government expenditures, financed in part by borrowing, often induce expansion in the money stock, as the Federal Reserve “monetizes” a part of the debt increase. Tax reductions also tend to expand the size of the deficit. The larger the deficit, the more likely is the Federal Reserve to increase its purchases of Treasury securities, which in turn, increases the money stock. 1964 1965 1966 1967 1968 1969 1970 1971 1972 Sources: U.S. Department of Commerce, Council of Economic Advisers, and Federal Reserve Bank of St. Louis Note: 1972 figures estimated by Federal Reserve Bonk of St. Louis and based on Federal budget for fiscal 1973. Latest d ata plotted: HEB-ith quarter preliminary; NIAB-4th quarter estimated The Federal budget deficit in fiscal year 1971 (on a unified budget basis) was $23 billion, the second largest deficit recorded since World War II. The budget deficit for fiscal year 1972, estimated at $11.6 billion in early 1971, has been revised upward to $38.8 billion. Reasons for the revision include an over estimate of the strength of the economy in calendar year 1971, and the Administration’s new fiscal pro posals of last August and this January. Congress has already acted favorably on a number of the Adminis tration’s proposals of August 1971. The high-employment budget, which assumes budget expenditures and tax revenues at a constant 4 percent level of unemployment, is expected to shift markedly from a $4.2 billion surplus (as estimated by this Bank) in fiscal 1971 to a $10.5 billion deficit in fiscal 1972. The fiscal 1973 high-employment budget is expected to run a $10.9 billion deficit. Thus the budget, by such actions as the 7 percent tax invest ment credit, the increased personal income tax exemp tions and accelerated expenditures, should have a stimulative effect on economic activity this calendar year. 1972 Projections Stimulative fiscal actions provide much of the basis for the very optimistic 1972 forecasts which have ap peared regularly in the media the past few months. Page 9 F E B R U A R Y 1972 F E D E R A L R E S E R V E B A N K OF ST. LO U I S A large majority of economic forecasters have pre dicted a strong surge in spending this year. Such unanimity is difficult to understand in view of the additional uncertainties with which the analysts are faced in 1972. The chief unknown influences on spend ing are usually monetary and fiscal actions, but this year the analysts must also project the expected im pact of the possible removal of some trade barriers, the dollar devaluation, and foreign exchange rate ad justments, as well as the effects of price-wage controls, their duration, and the successive phases, if any, of controls. Uncertainty often breeds divergence, but this year the product of uncertainty is conformity. The standard projections of economic activity in 1972 include: (1 ) approximately a $100 billion rise in total spending compared to a $73 billion increase in 1971; (2 ) an increase in real product growth from 2.7 percent in 1971 to about 6 percent; (3 ) a decline in the rate of price increase from 4.7 percent in 1971 to a little over 3 percent; and (4 ) a steady fall in the unemployment rate from 6 percent to a little over 5 percent by year end. Most forecasters believe these ebullient figures will be achieved by way of the following standard route: the consumer, bolstered by the progress of the wageprice control program and higher tax exemptions, starts spending more (and saving less); the increased expenditures reduce sellers’ inventories, which must then be replenished; a greater sales volume leads to higher profits which, together with the tax investment credit, induce capital expenditures; exports accelerate in response to increased demand from abroad, thereby creating many more jobs; residential construction and state and local spending pick up moderately, while Federal Government purchases of goods and services accelerate. MIT-PENN-SSRC M odel — Typical of the models which generate such projections for 1972 is the MITPENN-SSRC (MPS) econometric model. It is a large model of the economy with many behavioral relation ships and most of the latest features in model build ing, such as a well-developed financial sector. In addition, the model can give detailed projections of each economic sector in response to any of a large number of simulated policy actions. It, like most large models, incorporates cyclical forces, some structural changes, and possible policy alternatives. Unlike the small St. Louis Bank model, the larger, more detailed MPS model: (1) contains both money demand and supply functions; (2) can account to some extent for the expected improvement in U.S. net exports; (3) can simulate changes in activity Page 10 T a b le III M odel Projections of Economic Activity in 1972 Under Specific Assumptions* MPS Model 1971 A ctu a l St. Louis Model 1972 Projected 1972 Projected Rates of c h a n g e in: N o m in a l G N P .................... 7 .5 % 9 .4 % 7 .5 % Real G N P ......................... 2 .7 5 .8 3.5 Price D e f l a t o r .................... U n e m p lo ym e n t Rate . . . . 4 .7 3.3 3.9 5 9 5. 9 6. 2 * Among the many assumptions in the MPS model are a 6 percent rate of growth in the money stock, price and wage controls effec tive through III/1972 and rising prices of U .S. imports relative to exports. The St. Louis model assumes a 6 percent rate of growth in the money stock and high-employment expenditures as estimated by this Bank through IV/1972. Coefficients are estimated through 11/1971 for the MPS model and through III/1971 for the St. Louis model. under varying assumptions of both price and wage constraints; (4) has relatively long lags in the effect of changes in the money supply on economic activity and short lags in the effect of fiscal actions on economic activity; (5) projects estimates of total spending indirectly by the addition of GNP compon ents rather than directly. Table III indicates that the MPS model, under the assumptions that the money stock will grow at a 6 percent rate, that price and wage controls will be effective through III/1972, and that import prices will rise relative to the prices of U.S. exports, projects changes in total spending, real output, and prices in 1972 similar to those of the “standard” forecast.11 The unemployment rate projection is high relative to the standard. St. Louis M odel — Table III gives the projections of the St. Louis model for total spending, real output, prices, and the unemployment rate using the 6 percent money growth assumption employed with the MPS model. In this model, monetary and fiscal actions affect prices and real output (and, in turn, employ ment) by influencing the course of total spending. Monetary actions are the dominant source of changes in total spending. Whether the spending is channeled into real output changes or price changes depends on the degree of slack (actual output relative to poten tial) in the economy and the intensity of price antici pations. The model can account for cyclical forces, but offers a more limited selection of policy alterna tives than the large models. Some provision for struc tural change is found in the potential output variable. 11Six percent money supply growth was arbitrarily selected as the monetary assumption. Other assumptions are similarly arbitrary. F E D E R A L R E S E R V E B A N K OF ST. LO U IS The St. Louis model projects considerably smaller economic advances in 1972 over the year 1971 than either the standard or the MPS forecast (based on a 6 percent money growth assumption). Total spending, according to the St. Louis model, will rise 7.5 percent, the same as in 1971, real output will increase 3.5 percent, somewhat more than in 1971, prices will rise 3.9 percent, somewhat less than in 1971, and the unemployment rate will be little changed from this past year. There are three basic reasons why this set of St. Louis projections differs so much from the standard. First, the model has incorporated the sharply lower growth of the money stock in late 1971, the effect of which does not appear in the MPS or most other models for a long period. Second, no provision is made for the actions of the Price Commission and Pay Board. Third, the model does not account for the expected increase in foreign demand for U.S. goods. Because price and wage controls will likely keep price advances in 1972 below what they otherwise would be, and because increased demand for exports will probably stimulate spending, output and employ ment, the St. Louis projections for prices and un employment should be adjusted downward, and the projections for total spending and real output up ward, given a 6 percent rate of increase in the money stock. These projections should not be adjusted to the optimistic levels of the standard forecast, however, since the standard forecast is not appropriately “ad justed” for the influence of the recent slowdown in the growth rate of money. Looking Past 1972 To this point, only a limited set of figures for a few economic variables for a single year, 1972, have been discussed. To focus on such a narrow field is to miss much of the importance of current and future domestic economic developments. At least one issue — price and wage controls —merits further consideration. The adoption of price and wage controls by the U.S. Government in a period of peace time is a move which has strong long-term implications for our basi cally free market economy. Although there have been some brief periods of success with controls in this country as well as in a number of foreign countries, instances in which inflation was effectively curbed over sustained periods are rare indeed. Price and wage freezes have typically been followed by controls inequitably applied and ineffectively administered. The initial euphoria over the fact that someone is doing something to stop inflation has often given way F E B R U A R Y 1972 to dissatisfaction on the part of those whose incomes do not rise as fast as others and to cheating by those who cannot buy or sell goods at the administered prices. Once the controls are removed, past experience indicates prices may rise back to about where they would have been in the absence of controls.12 The MPS model, using the assumed 6 percent rate of monetary growth, indicates that prices would rise about as rapidly in 1973 as in 1972 if controls were removed toward the end of this year. This model also indicates that without any controls, but with moderate monetary growth, there would be less inflation in 1973 than in 1972, a result also given by the St. Louis model. Summary Nineteen seventy-one was a year of moderate re covery following the moderate recession in 1970. Com parisons with previous recession-early recovery years indicate that economic expansion in the 1970-71 pe riod was slightly weaker than in earlier comparable periods. A major difference is that prices rose at a much faster rate during the recent period than in the earlier ones. Some structural changes in the economy have undoubtedly occurred over the years, but they have probably played a minor role in influencing recent activity relative to normal cyclical adjustments and policy changes. Excessively stimulative monetary and fiscal policies over the extended 1965-68 period, for example, undoubtedly made the recent inflationary situation much more difficult to control than earlier ones. Most projections of economic activity in 1972 are quite optimistic relative to the actual experience of 1971. What makes 1972 that much different from 1971? The differences may be explored by examining cyclical influences, structural changes, and policy actions. The economy begins 1972 at only a slightly different position on the business cycle than a year ago in terms of actual “momentum” relative to potential. The '-L loyd Ulman and Robert Flanagan recently completed a study of incomes policies in other countries. “However, as indicated at the outset, periods of apparent effectiveness [of price-wage controls] were typically short-lived; they were frequently followed by wage or price explosions which sometimes blew up the policies themselves. Thus the policy at best seems to have been gaited for a short sprint rather than a long race, which suggests that it was better suited to deal with short-run emergencies like balance-of-payments disequilibria than with persistent inflationary forces.” See Lloyd Ulman and Robert J. Flanagan, W a g e R estraint: A Study o f In co m e P olicies in W estern E u ro p e (Berkeley and Los Angeles: University of California Press, 1971) p. 223. Page 11 F E D E R A L R E S E R V E B A N K OF ST. LO U IS rate of utilization of both labor and capital resources is little changed from this time last year. Actual output relative to potential output is approximately the same as a year ago. Some structural aspects of the economy such as the composition of the labor force and the ratio of goods to services will be little different than in 1971. Productivity may rise at a more rapid rate in 1972 than 1971, but short-run changes in productiv ity are more a result of cyclical changes than a cause. There are at least two institutional changes which will likely exert a strong influence on the economy in 1972. First, the anticipated improvement in demand for U.S. exports relative to imports due to exchange rate and tariff changes should have an expansionary effect on total spending, output and employment. Second, price and wage controls should contribute to a slowing in the rate of increase of prices in 1972. If the controls are effective over the short period of a year, many analysts believe this will have a positive influence on consumer confidence and lead to spend ing gains. The effect of controls on profits, work stop pages and price-wage escalator contracts is unclear. http://fraser.stlouisfed.org/ Page 12 Federal Reserve Bank of St. Louis F E B R U A R Y 1972 A major difference between 1971 and 1972 is that fiscal policy actions will likely be much more stimula tive. To the extent that the large, projected 1972 deficit is “monetized” by the Federal Reserve System, monetary policy may also be quite stimulative. Most economic analysts and models have not, however, ac counted for the recent slowing in the rate of growth of the money stock. Correction for this factor suggests somewhat lower projections than the norm for spend ing, output, and employment in 1972. Focusing on economic developments in 1972 runs the risk of overlooking the important long-range im plications of recent policy actions. The record of pricewage controls and sharp fluctuations in the money stock over a number of years gives little reason for continuation of such actions over an extended period. In developing stabilization policies, it should be recog nized that an economy the magnitude of ours cannot be steered in any direction instantaneously, and that effective policies should look to the long-run effects, rather than to the short cures which often lead to long-run instability. Operations of the Federal Reserve Bank of St. Louis — 1971 X HE ST. LOUIS Federal Reserve Bank and its branches provide numer ous services to member banks, the United States Government, and the public. These service operations include collecting checks, maintaining member bank reserve accounts, transferring funds, distributing coin and currency, and acting as fiscal agent for the Fed eral Government. The Bank also extends credit to its member banks, exercises supervision over certain commercial banks, engages in research directed to ward the formulation of monetary policy, and performs educational functions in regard to both monetary actions and Federal Reserve operations. In addition, it participates in Federal Open Market Committee deliberations, which are dis cussed in other issues of the R eview .1 T ab le 1 V O L U M E O F O P E R A T IO N S 1 D o lla r A m o u n t ( m illio n s) 1971 1970 Percent Change . . $ 1 5 5 , 8 0 3 .1 $ 1 3 8 ,9 4 5 .2 . . 2 6 8 .9 4 3 4 .3 C o in c o u n t e d ................................... . . 8 6 .0 7 9 .6 C urren cy c o u n t e d .............................. . . 1 ,8 6 6 .8 1 ,8 1 6 .2 2.8 T ran sfers of f u n d s .............................. . . 3 4 9 ,2 4 9 .4 2 8 7 ,4 6 7 .7 2 1 .5 B o n d s 3 ......................... . . 5 8 5 .0 5 7 9 .4 1.0 . . 2 3 ,6 2 9 .2 2 1 ,7 0 6 .2 8.9 22 9 .4 2 0 2.1 1 3 .5 C he cks collected2 N on cash collection U.S. S a v in g s O th e r G o ve rn m e n t .............................. items . securities3 . . . . U.S. G o ve rn m e n t c o u p o n s p a id . Num ber (th o u s a n d s ) 1971 1970 1 2 .1 % -3 8 .1 8 .0 Percent Change .............................. . . 4 8 0 ,9 4 6 3 9 7 ,5 0 4 N o n c a s h collection i t e m s .................... . . 804 855 c o u n t e d ................................... . . 7 2 5 ,8 8 5 7 1 5 ,5 0 5 1.5 c o u n t e d .............................. . . 2 4 8 ,8 0 6 2 4 0 ,4 5 2 3.5 T ran sfers of f u n d s .............................. . . 356 317 1 2 .3 C he cks C o in collected2 C urren cy 2 1 .0 % - 6 .0 This report on the operations of the — 2.0 . . 1 0 ,7 2 4 1 0 ,9 4 3 U.S. S a v in g s B o n d s 3 ......................... — 4 0 .6 O th e r G o ve rn m e n t securities3 . . . 581 978 Federal Reserve Bank of St. Louis in 1.2 804 814 U.S. G o ve rn m e n t c o u p o n s p a id . cludes the operations of its head office in St. Louis and its branches in Little 1Total for St. Louis office and the Little Rock, Louisville, and Memphis branches 2Excludes Government checks and money orders Rock, Louisville, and Memphis. The 3Issued, exchanged, and redeemed Federal Reserve Bank of St. Louis is one of twelve Federal Reserve Banks which, with Checks drawn on commercial banks are the means of settling the major portion of all nonbank financial their branches and the Board of Governors, form the Federal Reserve System. The St. Louis Bank operates transactions. Payment by check offers many advan tages over payment by cash, including less risk from in the Eighth Federal Reserve District, which includes theft, fire, and other disasters; greater convenience all of Arkansas and portions of Illinois, Indiana, Ken in making large transactions; greater opportunity to tucky, Mississippi, Missouri, and Tennessee. make large unscheduled transactions; and provision of a record of disbursements. The use of checks by Check Collection and Clearance individuals and businesses is facilitated by the collec The operation of the Federal Reserve Bank of St. tion and clearing operations of the Federal Reserve Louis utilizing the largest number of employees is Banks, which provide a mechanism for settlement of collecting and clearing checks. At the close of last checks collected by commercial banks. Settlement is year, 299 persons or 21 percent of the employees accomplished by entries to the member banks’ re at this Bank and its branches were engaged in the serve accounts at the Reserve Banks. check collection operation. The dollar volume of checks collected by the St. 'See Reprints 13, 17, 22, 28, 39, 57, and 68 for annual reviews Louis Federal Reserve Bank and its branches rose of monetary actions by the FOMC for the years 1964 through from $139 billion in 1970 to $156 billion in 1971, an 1970, available on request from this Bank. Page 13 F E D E R A L R E S E R V E B A N K OF ST. LO U IS increase of 12 percent. The number of checks col lected increased 21 percent from 398 to 481 million. These sharp rises in number and dollar volume of checks collected last year reflect improved economic activity (including inflation) in the district and a trend increase in the use of checks as a means of payment. The greater rise in the number of checks collected than in the dollar volume reflects the Bank’s new program of sorting many checks received from St. Louis area banks drawn on each other. Before mid-1970, these checks were sorted by the Suburban Check Exchange clearing banks and were cleared in bundles through the Collection Department. The dol lar value of checks cleared was not affected by this change, but the number increased because a pre sorted bundle is counted as one item, whereas checks sorted and cleared at this Bank are counted individually. Technological innovations, such as the conversion to electronic equipment, have increased the efficiency of collecting checks over the years. In 1960, prior to the installation of high-speed collection equipment, the St. Louis office and its three branches collected checks at the rate of 372 per man hour. By 1970 the rate of such collections had increased almost three fold to 1,026 checks per man hour. In 1971 approxi mately 97 percent of all checks received at this Bank were processed through computers. The Federal Reserve System in June 1971 outlined a program to streamline the structure of the U.S. payments system by expanding zones of overnight check clearance around Federal Reserve Banks and branches and by opening new regional zones for check clearance. The establishment of regional facili ties would accelerate check collections and increase availability of funds. The St. Louis Bank expanded its zones of immediate payment several decades ago to include most suburban banks in the St. Louis area. The zones of immediate clearing around St. Louis and its branch cities will be further expanded as transportation and equipment problems are resolved.2 In addition to collecting checks, the Federal Re serve Banks handle numerous other settlement items such as postal money orders, food stamp coupons, 2During the first half of 1972, the Little Rock immediate check clearing zone will be expanded to include Pulaski and Saline Counties in Arkansas; the Louisville zone will include Clark and Floyd Counties in Indiana and Jefferson County in Kentucky; and the Memphis zone will include Shelby County in Tennessee, Crittenden County in Arkansas, and DeSoto County in Mississippi. During the first half of 1973, the St. Louis zone will be expanded to include the City of St. Louis and St. Louis, Franklin, Jefferson, St. Charles and Warren Counties in Missouri, and Madison, St. Clair and Monroe Counties in Illinois. Digitized forPage 14 FRASER F E B R U A R Y 1972 and noncash collections. Among the latter are collec tions for drafts, promissory notes, bonds, and bond coupons. Transfer of Funds Wire transfer of funds is a service that the Federal Reserve provides to member banks to promptly and efficiently transmit funds around the country. These transfers primarily result from transactions involving interbank loans, check collections, and U.S. Treasury obligations. In implementing the new program for more rapid check clearance, the Federal Reserve in 1971 removed charges on wire transfers of $1,000 or more made through member banks by third parties — nonmember banks, businessmen and individuals — to encourage immediate payment of large amounts by wire. In addition, plans are being made to extend the hours of the wire transfer operations when warranted by a sufficient volume. The St. Louis Reserve Bank processed 356 thousand wire transfers amounting to $349 billion in 1971, an increase of 12 percent in number and 22 percent in dollar volume from 1970. These rapid increases in the number and dollar volume probably reflect in part the elimination of the above mentioned fees. Coin and Currency Operations The Federal Reserve Banks maintain a readily available supply of coin and currency to provide the amounts, kinds, and denominations demanded by the public. The public holds about a fifth of its money in currency, which is more universally acceptable than checks and is more desirable in settling some trans actions.3 The use of currency, especially for small transactions, is more timesaving and often a less ex pensive means of settlement than the use of checking accounts. To meet the public demand for currency, member banks order funds from their Reserve Bank, which charges the order to their reserve accounts. Nonmember banks generally receive their currency supplies from member banks. Those member banks with excess currency may deposit it in their reserve account at the Reserve Bank. The usable coin and currency is then redistributed, and the unfit is re moved from circulation to be melted down or destroyed. Coin and paper currency handled at the St. Louis Bank increased in 1971 as the demand for a hand-tohand medium of exchange rose with increased eco nomic activity. Pieces of coin counted and sorted 3Money is defined as demand deposits of the nonbank public plus currency outside banks. F E D E R A L R E S E R V E B A N K OF ST. LO UIS totaled 726 million, up 2 percent from 1970, and their value amounted to $86 million, up 8 percent. Pieces of paper currency counted and sorted increased 4 percent in 1971 to 249 million, and the value rose 3 percent to $1.9 billion. Fiscal Agency Operations As fiscal agents for the U.S. Treasury, the Federal Reserve Banks service the Treasury’s checking ac count, perform much of the work involved in the issuance and redemption of Government securities, and execute numerous other fiscal duties. The Federal Reserve Banks add to the Treasury’s checking account by periodically collecting Treasury funds from commercial banks. These funds were previously deposited in commercial banks by the pay ment of taxes and sales of Government securities. When the Government disburses its funds, the checks issued in payment are drawn against its account at the Federal Reserve Banks. When a new Government security is issued, the Federal Reserve Banks circulate subscription forms and receive applications for its purchase. The securi ties are allotted according to the Treasury’s instruc tions and are delivered to the purchasers. In addition, the Reserve Banks redeem Government securities upon maturity, make exchanges, and pay interest by redeeming coupons. In 1971 the four offices of this Bank issued, re issued, exchanged, and redeemed 11 million U.S. Sav ings Bonds valued at $585 million. The number de clined 2 percent from a year earlier, but the value rose one percent. The number of other Government securi ties issued, serviced, and redeemed decreased 41 percent in 1971 to 581 thousand while the dollar value rose 9 percent to $23.6 billion. The rapid de cline in the number of such Government issues processed, particularly with respect to securities is sued, resulted from several factors. After interest rates on short-term Government securities continued to decline in early 1971, funds of small investors were attracted to other investments, such as certificates of deposit, with higher rates of return. In addition, the Government no longer issued Treasury bills under $10,000 after March 1969. Therefore, no bills of a smaller denomination were handled after the matu rity dates in 1970. Lending Activity Member banks may borrow from Reserve Banks over short-term periods to meet their required re F E B R U A R Y 1972 serves. The volume of these borrowings tends to rise when the Federal Reserve lending rate is below other short-term rates, such as those on Treasury bills and prime commercial paper. Conversely, these loans decline when the discount rate, the interest rate charged to member banks on funds borrowed from Reserve Banks, is above these market rates. After being well below market rates during 1969 and most of 1970, the discount rate exceeded market interest rates during most of the past year. Reflecting these interest rate differentials, loans to member banks by the Federal Reserve Bank of St. Louis have declined sharply since 1969. Daily bor rowings averaged only $1.5 million in 1971, compared with $12.7 million in 1970 and $41.8 million in 1969. The percent of member banks borrowing from this Bank sometime during the year decreased from 17 in 1970 to 7 in 1971. The St. Louis Reserve Bank altered the discount rate six times during 1971 to realign it with fluctua tions in market rates. The effective rate on discounts under Sections 13 and 13a of the Federal Reserve Act was reduced from 5.5 percent to 5.25 percent on January 8, to 5 percent on January 29, and then to 4.75 percent on February 13. It was increased to 5 percent on July 16 and was reduced again on Novem ber 11 to 4.75 percent and on December 13 to 4.5 percent, the lowest rate since March 1968. In early 1971 the Board of Governors introduced several major changes in lending procedures to facili tate Federal Reserve credit services to member banks. The new procedures were designed to simplify dis count window accommodation by: 1 ) using a co n tin u in g len d in g agreem en t instead of an ap p licatio n and n ote fo r m ost borrow ings; 2 ) c o lle ctin g in te re st at th e tim e o f repaym ent ra th e r than b y d ed u ction in ad v an ce; and 3 ) m aking chang es in th e d iscou nt rate ap p licab le to outstand in g loans. Supervision Federal Reserve Banks perform a variety of super visory activities to foster effective operation of the commercial banking system. A major supervisory func tion is the annual examination of state chartered member banks to evaluate their assets, liabilities, capital, liquidity, operations, and management. This examination information is used by banking authori ties to correct unsatisfactory practices and conditions. Examiners from the Federal Reserve Bank of St. Page 15 F E D E R A L R E S E R V E B A N K OF ST. LO U IS Louis along with the state supervisory authorities examine the 111 state member banks in the Eighth Federal Reserve District.4 The staff of the Comptroller of the Currency examines the 347 national banks in the district, which are required by law to be mem bers of the Federal Reserve System. Other insured banks are examined by the Federal Deposit Insur ance Corporation and state supervisory authorities while the few noninsured banks are examined by state examiners only. Although fewer in number than the 1,064 nonmember banks, member banks hold about 60 percent of the total district bank deposits. Other supervisory functions of the Federal Reserve System include the admission of state banks to mem bership in the System and the approval of acquisitions by registered bank holding companies, bank mergers, and new branches of state member banks. During 1971, this Bank processed 6 applications to form onebank or multiple bank holding companies and 14 applications by holding companies to acquire addi tional subsidiaries. The passage of the “Bank Holding Company Act Amendments of 1970” placed onebank holding companies under Federal Reserve reg ulation similar to that of multiple bank holding com panies. These amendments will continue to increase supervisory activities. Approximately 75 of the esti mated 1,500 one-bank holding companies in the na tion have registered through this Bank. Research The research operations at the Federal Reserve Bank of St. Louis are directed toward the collection and analysis of regional, national, and international data. These data are instrumental in the formulation of monetary policy recommendations used by the President of this Bank in meetings of the Federal Open Market Committee. Information and data re lated to economic developments are available to the public in the monthly R eview and other periodicals. The Research Department’s role in reviewing pro posed bank holding company acquisitions and bank mergers has gready expanded in recent years. The passage of the “Bank Holding Company Act Amend ments of 1970” further increased research operations, especially with respect to one-bank holding companies. Bank Relations and Public Information This activity of the St. Louis Federal Reserve Bank is designed to maintain personal contact with all ^Number of banks as of December 31, 1971. Page 16 F E B R U A R Y 1972 banks in the Eighth District and to assist member banks with their operations relative to those of the Federal Reserve. Each year this department makes available to all district member banks the Federal Reserve Functional Cost Accounting Program, which provides a cost-income profile of each participating bank’s major functions. These data enable an in dividual bank to make comparisons with its previous operating statistics and with an average of banks of similar size. This department also distributes educational films on banking and makes arrangements for the display of currency exhibitions, speeches by the Bank’s staff, and tours of the Bank. Transfer of Counties from Eighth District Operations of the St. Louis Federal Reserve Bank in 1972 will be affected by the transfer of 26 member and 92 nonmember banks, located in 24 counties in western Missouri, from the St. Louis to the Kansas City Federal Reserve District. The twelve Federal Reserve Districts were originally constructed in 1914 by the Reserve Bank Organization Committee,5 which was authorized by the Federal Reserve Act to apportion the districts “with due regard to con venience and customary course of business.”6 In line with the efforts of the Federal Reserve to accelerate check clearances, the transfer of these counties, which are economically aligned with metropolitan Kansas City, will shorten the distances of check and cash delivery routes. This change, effective January 24, 1972, is the first in Federal Reserve district bounda ries since 1926 when two counties were transferred from the Dallas to the Kansas City District. Financial Statements Although the Federal Reserve Banks generate earn ings, their operations are directed primarily toward influencing monetary conditions. In 1971 the portion 5In making its districting decisions, the Organization Com mittee stated that it had been guided by the following six considerations: “mercantile, industrial, and financial connec tions existing in each district; ability of reserve bank in each district to meet legitimate demands on it; fair division of capital among the districts; general geographical and trans portation situation of the districts; population, area, and prevalent business activities of the district; and ability of member banks of each district to provide the minimum necessary capital.” See H. Parker Willis, The Federal Reserve System: Legislation, Organization and Operation (New York: The Ronald Press Company, 1923), p. 586. 6U.S., Congress, Senate, U.S. Reserve Bank Organization Committee, Location of Reserve Districts in the United States, 63rd Cong., 2nd Sess., 1914, document no. 485, p. 361. F E B R U A R Y 1972 F E D E R A L R E S E R V E B A N K OF ST. LO UIS of the System’s earnings allocated to the St. Louis Bank and its branches totaled $136.9 million, a decrease of one percent from a year earlier. This decline reflects the lower rates received on earning assets, which more than offset the increased holdings of such assets. After operating costs, net additions, statutory divi dends to member banks, and additions to surplus, the remaining earnings of $115.9 million were paid to the U.S. Treasury as interest on Federal Beserve notes. Dividends and transfers to surplus totaled $1.5 mil lion and $1.1 million, respectively. Total assets of the Federal Beserve Bank of St. Louis and its branches at the end of 1971 were $4 billion, an increase of 13 percent from a year earlier. Most of this rise resulted from increased holdings of Government securities, which are the principal means of creating Beserve Bank credit. Two-thirds of the assets were Government securities, primarily short term Treasury bills and notes. The remaining assets, including the gold certificate account, the special drawing rights certificate account, notes on other Be serve Banks, Federal agency obligations, cash items in process of collection, and bank premises, totaled $1.3 billion. Liabilities of the St. Louis Bank and its branches rose to $3.9 billion at the close of 1971. Federal Be serve notes, the principal type of currency in circula tion, amounted to $2.1 billion or more than half of this Bank’s liabilities. Deposits, consisting primarily of the reserve accounts of member banks, amounted to $1.2 billion, and other liabilities including deferred availability cash items and accrued dividends totaled $0.6 billion. T a b le III C O M P A R A T IV E STATEMENT OF C O N D IT IO N (D o lla r A m o u n ts in T h o u s a n d s ) A SSET S D ecem ber 31, 1971 G o ld Certificate A c c o u n t .................... 1 971 1970 Total e a r n i n g s ......................... $ 1 3 6 ,8 5 6 $ 1 3 8 ,7 3 8 N e t e x p e n s e s ......................... 2 1 ,9 1 2 1 5 ,0 0 0 1 5 ,0 0 0 4 0 ,1 4 0 3 2 ,1 0 2 O th e r C a s h ......................................... 1 7 ,4 7 5 1 2 ,2 7 9 Discou nts a n d A dvances . . . . A c c e p t a n c e s ......................................... Federal A g e n c y O b lig a t io n s . . . 1 1 4 ,9 4 4 1 2 0 ,4 9 0 3 ,5 5 1 — . 1 8 ,6 2 1 1 ,1 5 7 , 7 8 8 B i l l s .............................................. T ra n sferred to 1 ,4 7 4 1 ,2 1 8 ,8 1 0 1 2 6 ,1 7 5 1 0 7 ,8 2 5 T O T A L U.S. G O V E R N M E N T S E C U R I T I E S .......................... $ 2 , 6 4 9 , 0 1 9 $ 2 ,2 7 8 ,7 9 2 TOTAL L O A N S A N D S E C U R I T I E S .......................... $ 2 , 6 6 7 , 6 4 0 $ 2 ,2 7 9 ,1 7 2 C a sh Item s in Process o f C ollection . B a n k Prem ises 8 5 4 ,5 3 0 ( N e t ) ......................... 1 2 ,1 2 5 2 6 ,6 4 4 2 8 ,4 6 8 $ 3 ,9 8 2 ,3 1 9 $ 3 ,5 1 9 ,1 3 3 O th e r A s s e t s .................................... T O T A L A S S E T S ......................... L IA B IL IT IE S A N D 6 7 1 ,1 21 1 4 ,6 8 2 C A P IT A L A C C O U N T S L IA B IL IT IE S Decem ber 3 1 , 197 1 Federal Reserve N o te s $2,1 1 8 , 9 2 6 (N e t ) D ecem ber 31, 1 9 7 0 $ 1 , 9 5 1 ,2 2 1 D ep osits: . 1 ,0 1 5 ,1 7 8 8 8 4 ,7 6 1 G e n e ra l A ccou nt . 1 5 3 ,4 1 9 7 3 ,8 8 7 F o r e i g n ......................................... 9 ,5 2 0 4 ,2 5 0 2 7 ,8 5 1 1 0 ,9 0 6 TOTAL D E P O S I T S .................... $ 1 ,2 0 5 ,9 6 8 $ 1 2 0 ,9 0 6 $ 9 7 3 ,8 0 4 Deferred A v a ila b ilit y C a sh Items O th e r L iab ilities a n d A ccrued D iv id e n d s 2 2 ,1 1 2 2 0 ,9 7 2 T O T A L L I A B I L I T I E S .................... $ 3 ,9 3 1 ,9 6 7 $ 3 ,4 7 1 ,0 4 7 1 .4 % 20.1 - 5 8 4 ,9 6 1 7 5 3 .6 - C A P IT A L A C C O U N T S 2 5 ,1 7 6 . 2 4 ,0 4 3 2 5 ,1 7 6 2.0 2 4 ,0 4 3 — . $ 5 0 ,3 5 2 4.8 1 ,4 0 6 5 2 5 ,0 5 0 4 .6 C a p ita l P a id I n .................................... $ — 1 ,3 6 5 ,0 5 6 T O T A L C A P IT A L A C C O U N T S $ 9 5 2 ,1 5 7 — C e r t i f i c a t e s .................................... O th e r C a p ita l A c c o u n t s .................... Interest on Federal Reserve notes . — U.S. G o ve rn m e n t Securities: D istrib ution o f N e t E a rn in g s: D i v i d e n d s ......................... — — 416 N e t e a r n in g s before p aym en ts $1 1 8 ,4 9 5 to U.S. T re a su ry . . 380 — Percent Change 1 8 ,2 4 8 + 4 6 8 ,8 6 6 O lh e r D e p o s i t s ............................... LOSS STATEMENT ( D o lla r A m o u n ts in T h o u s a n d s ) N e t a d d itio n s ( + ) or d ed u ction s ( — ) . $ Federal Reserve N o te s o f O th e r B a n k s U.S. T re asu rer — C urrent net e a rn in g s . 3 4 6 ,2 0 8 S p e cia l D ra w in g Rights Certificate A c c o u n t ......................... M e m b e r B a n k ----Reserve A ccou nts T ob le II C O M P A R A T IV E PROFIT A N D $ Decem ber 31, 1970 — $ 4 8 ,0 8 6 T O T A L L IA B IL IT IE S A N D . . . surp lu s T o t a l .................... 1 1 5 ,8 8 7 11 8 ,2 9 8 - 2.0 1 ,1 3 4 1 ,2 0 2 - 5 .7 $ 1 1 8 ,4 9 5 $ 1 2 0 ,9 0 6 - 2 .0 % C A P IT A L A C C O U N T S . . . . $ 3 ,9 8 2 ,3 1 9 $ 3 ,5 1 9 ,1 3 3 MEMORANDA: Contingent liabilities on acceptances purchased for foreign correspondents increased from $8,503,000 on December 31, 1970 to $8,667,000 on December 31, 1971. Page 17 As of February 1, 1972 Directors Chairm an o f the Board and F ed era l R eserve Agent F r e d e r ic M. P e i r c e , Chairman and Chief Executive Officer, General American Life Insurance Company, St. Louis, Missouri D eputy Chairm an of the Board S a m C o o p e r , President, HumKo Products, Division of Kraftco Corporation, Memphis, Tennessee B r a d f o r d B r e t t , President, The First National Bank of E d w in S. J o n e s , Chairman and Chief Executive Officer, Mexico, Mexico, Missouri First National B a n k in St. Louis, St. Louis, Missouri I. B r o w n , J r ., President, Arkansas Foundry Company, Little Rock, Arkansas E dw ard President, Arkansas Bank and Trust Company, Hot Springs, Arkansas F red J. S c h n u c k , Chairman of the Board, Schnuck Markets, Inc., Bridgeton, Missouri J am es M. T u h o l s ic i , President, Mead Johnson & Company, Evansville, Indiana H a r r y M . Y o u n g , J r ., Farmer, Herndon, Kentucky C e c i l W . C u p p , J r ., LITTLE ROCK BRANCH Chairm an o f the Board m e l , Chairman of the Board, Southland Building Products Co., Little Rock, Arkansas J a k e H a r t z , J r ., President, Jacob Hartz Seed Co., Inc., A l P o l l a r d , President, A1 Pollard & Associates, Little Stuttgart, Arkansas Rock, Arkansas W. H K e l l e y President and Chief Executive Officer. £ llis £ ShelT0N5 President, The First National Bank of The State First National Bank of Texarkana, TexFayetteville. Fayetteville. Arkansas R oland R . R e m a r k a n a , A rk a n s a s E d w a r d M. P e n i c k , 3 President and Chief Executive Officer, Worthen Bank & Trust Company, Little Rock, Arkansas W A. S t o n e , Chairman and Chief Executive Officer, Simmons First National Bank of Pine Bluff, Pine Bluff, Arkansas ayne LOUISVILLE BRANCH Chairm an o f the Board President, Thomas Industries Inc., Louisville, Kentucky P a u l C h a s e , President, The Bedford National Bank, H e r b e r t J . S m i t h , President, The American National Bedford, Indiana Bank and Trust Company of Bowling Green, BowlH a r o l d E. J a c k s o n , President, The Scott County State ing Green, Kentucky Bank, Scottsburg, Indiana W i l l i a m H. S t r o u b e , Associate Dean, College of Science H u g h M. S h w a b , Chairman of the Boards, First Naand Technology, Western Kentucky University, Bowltional Bank of Louisville and The Kentucky Trust ing Green, Kentucky Company, Louisville, Kentucky ( vacan cy) J ohn G. B ea m , MEMPHIS BRANCH W il l ia m L. Chairm an o f the Board President, Mississippi State University, State College, Mississippi Gil e s , W ade C. B a r to n , President, First Citizens National Bank, Tupelo, Mississippi C. W h it n e y B rown , President, S. C. Toof & Company, Memphis, Tennessee J am es R. F itzh u g h , Executive Vice President, Bank of Ripley, Ripley, Tennessee Page 18 A l v in H uffm an, J r ., President, Huffman Brothers, Member, Federal Advisory Council Chairman of the Board and Chief Executive Officer, The Boatmen’s National Bank of St. Louis, St. Louis, Missouri D a vid H . M o r e y , In- corporated, Blytheville, Arkansas W ayne W . P ye a t t , President, National Bank of Commerce, Memphis, Tennessee J . J . W h it e , President, Helena National Bank, Helena, Arkansas Officers Darryl A. E ugene L eo n a ll C. W oodrow C. L. J o rd a n , E arl M. H. E dgar H. C r is t , R . Q u in n F o x , J . M . G e ig e r , W. Assistant Vice President Assistant Counsel and Assistant Secretary of the Board K a th r yn J . M o re, W il l ia m R. M u eller, A lexan d er P . Orr, Assistant Vice President Assistant Vice President Assistant Vice President P a u l Salzm an , R o b e r t W . T h o m a s, K arl Assistant Vice President E. Assistant General Auditor Assistant Vice President J o h n F . O t t in g , J r ., Assistant Vice President Assistant Vice President V iv ia n , Assistant Vice President Assistant Vice President Assistant Chief Examiner Assistant Vice President K er a n , Assistant Vice President C l if t o n B . L u t t r e l l , Chief Examiner R ic h a r d 0 . K a l e y , M ic h a e l Vice President Assistant Vice President W il l is L . J o h n s, Senior Vice President Senior Vice President Chief Examiner J o h n W . D r u e l in g e r , e ig e l , Vice President Vice President F . G a r l a n d R u s s e l l , J r ., Vice President, General Counsel, and Secretary of the Board C h a r l e s E. S i l v a , Vice President H a r o l d E. U t h o f f , Vice President Assistant Vice President G e o r g e W . D e n n is o n , W D o n a ld W . M o r i a r t y , J r ., Vice President* Ca r lso n , C h a p in , H. J o h n W . M en g es, General Auditor N orm an N. B o w s h er , K e it h otaw a, Vice President G e o r g e W . H ir s h m a n , J erry W Vice President W . Gil m o r e , President First Vice President H ow ard J o seph J o s e p h P . G a r b a r in i , F r a n c is , Senior Vice President A nd ersen , Ger a ld T . D u n n e, R. L eo n a r d , D elm er D. W Assistant Vice President** C h arles D. Z et t le r , e is z , Assistant Chief Examiner LITTLE ROCK BRANCH Jo h n F. B reen , Vice President and Manager M i c h a e l T . M o r i a r t y , Assistant Vice President and Assistant Manager Thom as R. C a lla w a y , Assistant Vice President Jo h n K. W a rd , Assistant Vice President LOUISVILLE BRANCH D on ald L . H e n r y , J am es R o bert E . Harlo w , E. C on rad, Senior Vice President and Manager Assistant Vice President and Assistant Manager Assistant Vice President G e o r g e E . R e i t e r , J r ., Assistant Vice President MEMPHIS BRANCH L. P aul R u th A . B ryant, I. B la ck , T e r r y B r it t , Vice President and Manager Jr., Assistant Vice President and Assistant Manager Assistant Vice President A n t h o n y C. C r e m e r iu s , Jr., Assistant Vice President * 0 n leave to Bundesbank, Germany. * * 0 n leave to the U.S. Treasury Department, Washington, D.C. Page 19 Projecting With the St. Louis Model: A Progress Report by KEITH M. CARLSON T h e ST. LOUIS model was designed to project the general time path of response of certain key economic variables to monetary and fiscal actions.1 What can be said, in retrospect, about the success of the model in achieving its stated objectives? The period from fourth quarter 1969 to second quarter 1971 is taken as the focus of discussion. Late 1969 is chosen as the starting point for the analysis because the current version of the St. Louis model was developed at that time. Mid-1971 is selected as the terminal point of reference because of the adop tion of an incomes policy in mid-August. The model, as designed, is not able to capture the short-run effects of price-wage controls, nor does it allow in corporation of the effects of restructured international monetary arrangements. The model is purposely kept small to aid in the identification of the fundamental determinants of economic trends over periods as long as a year or more. The St. Louis model is small in size and is not designed for quarter-to-quarter forecasting because many factors are known to influence short-run move ments in economic activity. For this reason, the model’s performance is examined for the six-quarter period as a whole, rather than quarter by quarter. There are three exogenous variables in the St. Louis model — changes in the money stock and high em ployment Federal expenditures, considered sum mary measures of monetary and fiscal actions, and potential output, which reflects growth of the labor force and productivity.2 In using the model for purposes of monetary policy recommendation, alter1Leonall C. Andersen and Jerry L. Jordan, “Monetary and Fiscal Actions: A Test of Their Relative Importance in Economic Stabilization,” this Review (November 1968), pp. 11-24, and Leonall C. Andersen and Keith M. Carlson, “A Monetarist Model for Economic Stabilization,” this Review (April 1970), pp. 7-25. 2For a discussion of the role of potential output in the St. Louis model, see Roger W. Spencer, “Population, Labor Force, and Potential Output: Implications for the St. Louis Model,” this Review (February 1971), pp. 15-23. Page 20 native steady growth rates of the money stock are assumed. This procedure does not entail forecasting movements in the money stock over short periods. Instead, it presents a set of simulations using alterna tive steady growth rates which can aid in assessing the economic impact over several quarters of different trend growth rates of money. Background The economy was in the early stage of a recession in late 1969.3 GNP had risen 7.5 percent in 1969, compared with 8.9 percent the previous year, and the rate of increase from third to fourth quarter of 1969 had slowed to 3.4 percent. Real product had declined from third to fourth quarter 1969. Despite a slowdown of growth in nominal and real terms, prices had continued to accelerate through 1969. St. Louis model projections in early 1970 indicated there was little prospect of strong economic recovery in 1970 because of the lagged effect of monetary restraint in 1969.4 For example, assuming 6 percent growth in money, real output was projected to de cline in the first two quarters of 1970, pick up slightly in the second half, then advance at a 3.5 to 4 percent rate in 1971. The outlook for prices with a 6 percent rate of monetary growth was for a very slow decline in the rate of inflation throughout the projection period — from a 4.7 percent rate of increase in fourth quarter 1969, to a 4.3 percent rate in fourth quarter 1970, and a 3.8 percent rate in late 1971. Unemployment was projected to rise quite rapidly in 1970, then level off in 1971 at about 5.7 percent. The model presented a mixed picture for interest rates. Long-term rates were projected to change 3See “Real Economic Expansion Pauses,” this Review (Febru ary 1970), pp. 2-7. 4See Andersen and Carlson, “A Monetarist Model,” p. 19. F E D E R A L R E S E R V E B A N K OF ST. LO U IS C h a rt I Data Used in Calculations of Policy Assumption Error . _ P o lic y V a r i a b l e s * . _ A n n u a l R ates A n n u a l R ates of C h an g e Money Stock of Change ------------------------ ------------------------ ------------------------ 15 15 H i g h - E m p lo y m e n t Budc 'ixpenditures F E B R U A R Y 1972 prepared. One simulation was based on coefficients estimated with data for the variables of the model through second quarter 1971, and used observed values for money and Federal expenditures to drive the model in the simulation period. This simulation is referred to as the “ex post” simulation. Another simu lation, called “ex ante A,” used coefficients estimated from data through fourth quarter 1969, and like the “ex post” simulation, actual data on money and Federal expenditures were used to drive the model. Finally, a simulation referred to as “ex ante B ” was run which again had coefficients based on data through late 1969, but steady growth rates of money and Federal expenditures (at their average rates over the simula tion period from late 1969 to mid-1971) were used. These three simulations were designed to identify the sources of error underlying model projections based on constant growth rates of money and Federal expenditures during the six-quarter period.5 Total er ror is defined as the difference between the observed value (either its level or rate of change) of a variable and the value projected in the ex ante B simulation. These values for the variables in the ex ante . B simulation can be compared with those published in the April 1970 issue of this Review. For purposes of evaluation, total projection error is divided into explained and unexplained error. Unexplained error is the difference between actual values and those yielded by the ex post simulation. Explained error is attributed to two factors —chang ing economic structure, which is reflected in changes in the coefficients, and deviations of monetary and fiscal variables from the assumed steady rates. The method used here to identify the sources of error is summarized as follows: Error Defined as: Actual minus Ex Post Ex Post minus Ex Ante A 1969 1970 Due to changing eco nomic structure Ex Ante A minus Ex Ante B Due to steady rate assumption Actual minus Ex Ante B Total error 1971 * S h a d e d a r e a s r e p r e s e n t s im u la t io n p e r io d s . little for the period through 1971, while short term rates were projected to decline sharply in 1970, to about 6.3 percent, then drop further to 5.5 per cent by late 1971. Procedure To provide a basis for a systematic evaluation of the St. Louis model, three sets of simulations were Unexplained 5The sources of error relate to the use and interpretation of the St. Louis model, and are not to be confused with studies of sources of forecasting error published by other investigators. See, for example, Jared J. Enzler and H. O. Stekler, “An Analysis of the 1968-69 Economic Forecasts,” The Journal of Business (July 1971), pp. 271-81, for a dis cussion of forecast error attributable to inaccurate predictions of public policy. Page 21 F E B R U A R Y 1972 F E D E R A L R E S E R V E B A N K OF ST. LO U I S Unexplained error provides an indication of the overall reli ability of the specification of the model. Error attributable to changing economic structure provides a measure of the sta bility of the coefficients over time, although with the addi tion of only six quarters of data, this is a very weak test of para meter stability. Finally, error attributable to the steady rate assumption regarding the mone tary and fiscal variables sheds light on the usefulness of the model as a guide to the formu lation of policy. Results The results of the three types of simulations with the St. Louis model for first quarter 1970 through second quarter 1971 are summarized in Table II and Chart II. The sources of error as defined above are shown in Table I. The focus of discussion is the average error for each variable, i.e., the last co lu m n of Table I. Rates of change in total spending, real product and prices are emphasized because the part of the model relating to these variables is estimated in first dif ference form, reflecting greater confidence in the statistical reli ability of first difference esti mates in the GNP accounts than of levels. T a b le I SO U R C ES OF ERROR IN ST. LOUIS M O D E L* (Percent) 1970 1 — II 1971 III IV 1 II Q u a rte r A ve rage Total S p e n d in g -2 .7 -1 .7 -2 .0 — 6 .8 5 .5 -0 .5 -1 .3 7 U n e x p la in e d error -1 .6 -2 .0 -0 .5 — 4 .0 7 .7 -1 .5 -0 .3 2 Error d u e to c h a n g in g econom ic structure — 0 .2 -0 .3 -0 .5 -0 .3 0.5 0 .5 — 0 .0 5 Error d u e to ste a d y rate assu m p tio n -0 .9 0 .6 -1 .0 -2 .5 -2 .7 0 .5 -1 .0 0 -3 .8 -1 .2 -1 .9 -7 .7 4 .7 0 .2 -1 .6 2 — 2 .4 — 1.1 -0 .1 -4 .7 7.1 -0 .5 -0 .2 8 Error d u e to c h a n g in g econom ic structure -0 .6 -0 .7 -0 .9 -0 .6 0 .0 0 .0 -0 .4 7 Error d u e to ste a d y rate a ssu m p tio n -0 .8 0 .6 -0 .9 -2 .4 -2 .4 0 .7 — 0 .8 7 Total error Real Product Total error U n e x p la in e d error Prices Total error 1.5 -0 .5 0.1 1.3 0 .4 -0 .7 0 .3 5 1.1 -0 .9 -0 .3 1.0 0 .2 -0 .9 0 .0 3 Error d u e to c h a n g in g econom ic structure 0.4 0 .5 0 .4 0 .4 0 .5 0 .4 0 .4 3 Error d u e to ste a d y rate assu m p tio n 0 .0 0 .0 — 0.1 -0 .3 -0 .2 -0 .1 2 U n e x p la in e d error -0 .1 U n e m p lo ym e n t Rate -0 .3 0 .0 0 .3 0 .9 0 .8 0 .9 0 .4 3 -0 .3 — 0.1 0.1 0 .6 0 .3 0 .2 0 .1 3 Error d u e to c h a n g in g econ om ic structure 0 .0 0.1 0.1 0.2 0 .3 0 .2 0 .1 5 Error d u e to ste a d y rate a ssu m p tio n 0 .0 0 .0 0.1 0.1 0.2 0 .5 0 .1 5 0 .7 0 .8 0.8 0 .3 -0 .4 -0 .2 0 .3 3 0 .3 0 .5 0 .5 0 .0 -0 .5 0 .0 0 .1 3 Error d u e to c h a n g in g econom ic structure 0 .3 0 .3 0 .2 0 .3 0 .2 0 .2 0 .2 5 Error d u e to ste a d y rate a ssu m p tio n 0.1 0 .0 0.1 0 .0 -0 .4 -0 .0 5 Total error U n e x p la in e d error C o rp o ra te A a a Rate Total error U n e x p la in e d error -0 .1 C om m ercial P a p e r Rate 1.5 Total error U n e x p la in e d error Error d u e to c h a n g in g econ om ic structure Error d u e to ste a d y rate a ssu m p tio n 1.1 0 .6 — 0 .9 -2 .5 — 1.9 -0 .3 5 1.1 1.1 0 .6 -0 .8 -1 .7 -0 .5 -0 .0 3 0.1 0 .0 -0 .2 -0 .2 -0 .1 -0 .0 8 -0 .1 Total spending — Assuming -0 .2 3 0 .0 0.1 0.1 -0 .6 -1 .3 0 .3 steady growth of money and ex *Total error = actual — ex ante B penditures at the average rate Unexplained error = actual — ex post E rro r due to changing economic structure = ex post — ex ante A that actually occurred (ex ante E rro r due to steady rate assumption = ex ante A — ex ante B B simulation), the average an specification of the St. Louis model, average projection nual rate of change of total spending would have been error was 0.32 percentage points for this period. overpredicted by 1.37 percentage points. Of this over prediction, 0.32 is unexplained. In other words, using The average error associated with changing eco all available information, including observations in nomic structure is .05 percentage points. Updating the simulation period, but retaining the fundamental Page 22 F E B R U A R Y 1972 F E D E R A L R E S E R V E B A N K OF ST. LO U IS Errors in U n ex p la in e d the St. Louis Model: 1/1970 D ue to Changing Structure -----A C TU A l n - Due to Steady G row th Assumption ----- EX POST U. ------- EX ANTE * ------- EX POST ------- EX ANTE * ------- EX ANTE B A n n u a l R ates of C h a n g e u 1969 A n n u a l Rates of C h an ge 1 969 A n n u a l R ate s of C h a n g e 11 A n n u a l Rates of C h a n g e 1 969 A n n u a l R ates of C h a n g e A n n u a l Rates of C h a n g e 11/1971 1969 A n n u a l R a te s of C h a n g e ♦ E ffects o f m a jo r s trik e ' 1969 n u a l Rates Change ♦E ffects o f m a jo r strike 1970 1970 UMwptotmmt Rift P erce nt Unemployment Rile P e rtsn l Corporate Ah Rile 1970 1970 Cwpwite Am Rite Uweiployiiieiit Kite P erce nt Corporile Am Rale P ercent rcent P e rce n t Commercial Piper Rale 1970 1970 1970 1969 197 0 IX A ll p a n e ls show a c tu a l d a ta fo r 1969. Page 23 F E B R U A R Y 1972 F E D E R A L R E S E R V E B A N K OF ST. LO U IS the coefficients with six addi tional quarters of data improves the accuracy of the total spend ing projection by .05 percentage points, on average. T a b le II ST. LO U IS M O D E L S IM U L A T IO N S V S. A C T U A L (Percent) 1970 1971 1 The average error for total spending attributable to the as sumption of steady growth of the policy variables is the larg est, amounting to one percent age point. This result follows from the pattern of variability of quarter-to-quarter rates of change of both money and ex penditures during this six-quarter period (see Chart I). The annual rate of change of money varied from 4.2 to 11.3 percent to yield the 6.5 percent average rate of change for the period, whereas Federal expenditures varied between — and 24 per 3.4 cent, averaging 8.6. II III IV 1 II S ix Q u a rte r A ve rage 6 .4 2 ACTUAL A n n u a l Rates o f C h a n g e in: 3 .4 5 .3 6 .3 2 .0 13.8 7 .7 — 2.9 0 .7 1.2 - 4 .0 7 .9 3 .4 1 .0 5 6 .6 4 .6 5.1 6 .3 5 .4 4 .2 5 .3 7 U n e m p loym e n t Rate 4 .2 4.8 5 .2 5 .9 5 .9 6 .0 5 .3 3 C o rp o ra te A a a Rate 7.9 8.1 8.2 7 .9 7.2 7 .5 7 .8 0 C om m ercial P a p e r Rate 8.6 8.2 7.8 6 .3 4.6 5 .0 6 .7 5 6 .7 3 Total S p e n d in g Real Product Prices EX P O S T S IM U L A T I O N 1 A n n u a l Rates of C h a n g e in: 5 .0 7 .3 6.8 6 .0 6.1 9.2 -0 .5 1.8 1.3 0 .7 0 .8 3 .9 1 .3 3 5 .5 5 .5 5 .4 5 .3 5 .2 5.1 5 .3 3 U n e m p loym e n t Rate 4 .5 4 .9 5.1 5 .3 5 .6 5 .8 5 .2 0 C o rp o ra te A a a Rate 7 .6 7 .6 7 .7 7 .9 7 .7 7.5 7 .6 7 C om m e rcial P a p e r Rate 7 .5 7.1 7 .2 7.1 6.3 5 .5 6 .7 8 Total S p e n d in g Error attributable to the steady rate assumption would not nor mally be expected to be this large, but it is a factor to con sider when using this equation for a horizon as short as six quarters. It is not only the vari ability but also the pattern of the variability that leads to this type of error. For example, con sider the accompanying figure on the following page, where two patterns of money growth are shown, both of which yield the same average for the period. Total S p e n d in g 5 .2 7 .6 7 .3 6 .3 5 .6 8 .7 6 .7 8 Real Product 0.1 2.5 2.2 1.3 0 .8 3 .9 1 .8 0 Prices 5.1 5 .0 5 .0 4 .9 4 .7 4 .7 4 .9 0 U n e m p lo ym e n t Rate 4.5 4.8 5 .0 5.1 5 .3 5 .6 5 .0 5 C o rp o ra te A a a Rate 7 .3 7 .3 7 .5 7 .6 7.5 7 .3 7 .4 2 C om m ercial P a p e r Rate 7 .4 7.1 7 .3 7 .3 6 .5 5 .6 6 .8 7 Total S p e n d in g 6.1 7 .0 8.3 8.8 8.3 8.2 7 .7 8 Real Product 0 .9 1.9 3.1 3 .7 3 .2 3.2 2 .6 7 Prices 5.1 5.1 5 .0 5 .0 5 .0 4 .9 5 .0 2 Real Product Prices EX A N T E A S IM U L A T I O N 2 A n n u a l Rates o f C h a n g e iin: EX A N T E B S IM U L A T I O N 3 A n n u a l Rates o f C h a n g e i n: U n e m p lo ym e n t Rate 4 .5 4.8 4 .9 5 .0 5.1 5.1 4 .9 0 C o rp o ra te A a a Rate 7 .2 7.3 7 .4 7 .6 7 .6 7 .7 7 .4 7 C om m ercial P a p e r Rate 7.1 7.1 7 .2 7.2 7.1 6 .9 7 .1 0 'Coefficients Coefficients Coefficients simulation estimated through 11/1971 ; actual money and expenditures used in simulation. estimated through IV/1969 ; actual money and expenditures used in simulation. estimated through IV /1969; money at 6.5 percent and expenditures a t 8.6 percent in period. (Averages of actual quarter-to-quarter rates from 1/1970 to 11/1971.) Because of the lags in the impact of monetary actions, the money growth pattern in Case B will be associated with a larger growth of total spending for the period than in Case A.6 If total spending growth were con sidered for a longer period of time, the pattern would be irrelevant. Furthermore, the error in total spending attributable to the steady rate assumption can be traced primarily to the pattern of variation in monetary growth, rather than Federal expenditures, because of the nature of the coefficients in the total spending equation. Page 24 6Money growth in 1971 is a case in point. Assuming lags in the impact of monetary actions, as used in the St. Louis model, the steady rate equivalent is calculated as follows: Weighted Annual Rate Annual Rate of of Change Change of Money Weight 2.74 7.3 .375 1/71 3.62 .320 11/71 11.3 .214 1.65 111/71 7.7 .090 .04 .4 IV/71 8.05 An unweighted average of the monetary tgrowth rates would be 6.7 percent. The economic impact of monetary actions in 1971, though averaging a 6.7 percent growth, is the equi valent of an 8 percent steady rate in the St. Louis model. F E D E R A L R E S E R V E B A N K OF ST. LO UIS A lte r n a tiv e P atterns o f M o n e y G r o w th C a se B C a se A P e rc e n t P e rce n t F E B R U A R Y 1972 product projections bear the brunt of two types of errors — that due to the steady rate assumption in the total spending projections and that due to chang ing economic structure in the price projections. Unemployment rate — Projections of the unemploy ment rate can be expected to reflect the projection errors for real product. The total error associated with the unemployment rate averaged 0.43 percentage points. The total was equally divided among the three categories of error. Tim e Prices — The total error for price projections was 0.35 percentage points during the six-quarter sim ulation period. However, unexplained error was negligible. The major source of error in the price projections was associated with changing economic structure. Up dating the price equation with observations from the six-quarter simulation period alters the coefficients significantly —in particular, increasing the coefficient on the weighted sum of past prices. The implications of this result are unclear, since our knowledge of the determinants of price level movements and price ex pectations and their interaction is so limited. Error attributable to the steady rate assumption was in a direction opposite that due to changing eco nomic structure. In other words, the steady growth case (ex ante B ) tended to predict faster inflation than when the actual course of the policy variables (ex ante A) was included in the ex ante simulations. Like total spending, the error for prices attributable to the steady rate assumption can be traced to the pattern of variation in monetary growth. R eal product —The projections of real product are dependent on those for total spending and prices because real product is calculated by subtracting from the change in total spending that portion which is associated with estimated price change. Overpredic tion of total spending along with underprediction of prices leads to an average overprediction of real product of 1.62 percentage points. Unexplained error is about the same as for total spending because such error is negligible for prices. Changing economic structure accounts for a sub stantial portion of total error for real product, re flecting the underprediction of prices. Error due to the steady rate assumption is also significant and of the same order of magnitude as for the total spending projections, though it does receive some correction from such error in the price projections. The real It should be noted that the sources of error are difficult to identify for the unemployment rate be cause the errors are derived from full model simula tions rather than the unemployment equation alone. All types of error come into play, operating through total spending and prices, as well as the unemploy ment rate equation itself. C orporate Aaa rate — Examination of the corporate Aaa rate projections indicate total error averaged 33 basis points. Unexplained error averaged 13 basis points, but changing economic structure is the key factor in the Aaa projections. This result is to be expected because price movements play an important role in the corporate Aaa equation. Because prices tended to be underpredicted in the ex ante A simula tion, the Aaa rate is underpredicted also. There is a partial offset from the overprediction of real product growth, but clearly the price effect dominates. Com m ercial paper rate — Commercial paper rate projections exhibit an average total error of about the same absolute magnitude as the corporate Aaa pro jection, but, in contrast, it is overpredicted. Surpris ingly, both unexplained error and that due to chang ing economic structure are small during the simulation period. The reason is that the overprediction of real product and the underprediction of prices tended to offset each other. The price effect does not dominate the short-rate projections like it does the long-term rate. Error attributable to the steady rate assumption accounts for most of the total error in the commercial paper rate projections. One of the reasons is that the direct effect of money tends to be larger in the short-rate equation than for the long rate. In addition, over-predictions of both real product and prices, as associated with the steady rate assumption effect, reinforce each other in the commercial paper rate projections. Conclusions The St. Louis model was not designed for quarterto-quarter forecasting. When interpreted in light of Page 25 F E D E R A L R E S E R V E B A N K OF ST. LO UIS F E B R U A R Y 1972 its objectives, the model succeeded in roughing out the average time paths of total spending, real prod uct, prices, unemployment, and interest rates during the period from late 1969 to mid-1971. However, the degree of success can be gauged only by comparing the results with those of other models.7 7Some general comparisons with the Wharton model and the FRB-MIT-Penn model are given in Lawrence R. Klein, “Empirical Evidence on Fiscal and Monetary Models,” in James J. Diamond, ed., Issues in Fiscal and Monetary Policy: Policymakers and private forecasters may find that the St. Louis model is inadequate for many of their purposes—for example, quarter by quarter forecasts or detailed forecasts of the components of GNP. Never theless, it is hoped that this identification of strengths and weaknesses of the model will provide further understanding on its use and interpretation. The Eclectric Economist Views the Controversy (DePaul University, 1971), pp. 35-50. APPENDIX Estimated Equations of the St. Louis Model' I. A. Total Spending Equation A. Sample period: Sample period: .66 3 .8 4 1.80 Sample period: Sample period: .92 .30 .65 1 /1 9 5 5 — 11/1971 .58 4 .5 3 2.12 A. 1 /1 9 5 5 — IV /1 9 6 9 Sample period: 1 /1 9 5 5 — IV /1 9 6 9 R1 = 1.23 ,' .06 M , + 1.44 Z , + .21 X t_, (4 .9 0 ) ( - 3 . 5 2 ) (1 1 .3 6 ) (2 .9 6 ) .87 1.11 1.39 + .99 P / ( U / 4 ) , _ , (1 8 .5 1 ) R2 = S.E. = D -W = 1 /1 9 5 5 — 11/1971 APt = 2 .0 5 + .08 D t_ , + 1.01 A Pt A (6 .3 2 ) (8 .3 8 ) (1 5 .7 0 ) R2 .90 S.E. = 1.23 D -W = 1.67 “Constraints and lag structures correspond to those set forth in the original article discussing the St. Louis model. See Ander sen and Carlson, “A Monetarist Model.” Coefficient values on lagged variables (subscripted “t-i” ) are sums of the coeffi cients for current and lagged quarters. Figures enclosed by parentheses under the coefficients are “t” statistics. Page 26 Sample period: IV. Long-Term Interest Rate Equation APt = 2 .4 6 + .09 D t_ , + .93 AP,A (6 .2 0 ) (8 .6 4 ) (9 .0 4 ) R2 = S.E. = D -W = B. R2 = S.E. = D -W = B. .92 .31 .59 U , = 3 .8 8 + .03 G, + .29 G ,- , (7 2 .4 1 ) ( .8 3 ) (7 .6 6 ) II. Price Equation A. 1 /1 9 5 5 — IV /1 9 6 9 R2 = S.E. = D -W = 1 /1 9 5 3 — 11/1971 AY, = 2 .7 0 + 5.11 AM t_ , + .09 A E ,_ , (3 .0 4 ) (7 .3 7 ) (.2 5 ) R2 = S.E. = D-W = Sample period: U , = 3 .8 9 + .04 G, + .29 G t_ , (7 0 .5 9 ) (.8 7 ) (6 .9 4 ) 1 /1 9 5 3 — I V / 1969 AYt = 2.78 + 5 .1 0 A M ,., + .10 A E ,_ , (.3 0 ) (3 .6 1 ) (7 .9 9 ) R2 = S.E. = D-W = B. III. Unemployment R ate Equation B. Sample period: .92 .28 .67 1 /1 9 5 5 — 1 1 /1971 R , = 1.29 .06 M t + 1.58 Z , + .15 X , - , (4 .7 7 ) ( - 3 . 5 6 ) (1 3 .0 0 ) (2 .1 2 ) + 1.03 P / ( U / 4 ) , _ , (1 9 .9 6 ) R2 = S.E. = D -W = .95 .31 .79 F E D E R A L R E S E R V E B A N K OF ST. LO U IS V. F E B R U A R Y 1972 S h o rt-T erm In terest R a te E q u a tio n A. Sample period: AP = AY - XF = potential output X + 1.30 P / ( U / 4 ) t_, (1 5 .1 7 ) R= S.E. = D -W = Sample period: D = output (G N P in 1958 prices) 1 /1 9 5 5 — IV /1 9 6 9 Rf = - 1 .0 7 .17 M t + 1.01 Z t + .66 X t_ , ( —2 . 7 2 ) ( —5 .6 7 ) ( 5 .1 7 ) (7 .3 3 ) B. = dollar change in total spending (G N P in current prices) due to price change .88 .51 .54 (X F- X ) A PA = anticipated price change (scaled in dollar units) Rf = - 1 . 0 0 .18 M t + 1.05 Z t + .64 X t_, ( —2 .1 9 ) ( —5 .2 1 ) (4 .7 3 ) (6 .0 1 ) = unemployment as a percent of labor force G 1 /1 9 5 5 — 11/1971 U = R ((X F - X )/X F) • 100 Moody’s seasoned corporate Aaa bond rate M = annual rate of change in money stock Z = dummy variable ( 0 for 1 /1 9 5 5 — IV /1 9 6 0 and 1 for 1 /1 9 6 1 — end of regression period) X = annual rate of change in output (G N P in 1958 prices) AY = dollar change in total spending (G N P in current prices) P = annual rate of change in GNP price deflator (1 9 5 8 = 1 0 0 ) AM = dollar change in money stock U / 4 — index of unemployment as a percent of labor force (base = 4 .0 ) + 1.30 P / ( U / 4 ) t_i (1 4 .3 3 ) R= S.E. = D -W = .87 .60 .61 Symbols are defined as: AE = dollar change in high-employment Federal expenditures Rs = four- to six-month prime commercial paper rate Page 27 SU B SC R IPTIO N S to this Bank’s R e v ie w are available to the public without charge, including bulk mailings to banks, business organizations, educational institutions, and others. For information write: Research Department, F ederal Reserve Bank o f St. Louis, P. O. Box 442, St. Louis, Missouri 63166.