The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
For release on Friday, Apr. 21, 1967 at:12 :30 p.m. EST Monetary Policy and Municipal Finance Remarks of George W. Mitchell Member, Board of Governors of the Federal Reserve System at the Annual Seminar of the California Municipal Treasurers' Association Anaheim, California April 21, 1967 Monetary Policy and Municipal Finance Superficially, municipal finance officers and central bankers would not seem to have much in common. But it is not inappropriate that I should be here speaking about monetary policy, for monetary policy affects interest rates and interest rates affect your business. And to be more specific, your response to changing interest rates has something to do with the efficiency with which you manage your governments' finances. No doubt there are many occasions on which you have been puzzled or exasperated at the trend of monetary policy because of the uncertainty and frustration it has introduced into your capital spending plans and financing. I should confess that there are occasions on which 1 have been puzzled and exasperated at your behavior, too. Why is it that you insist on embarking on municipal projects of less, as well as greater, priority when everyone else in the nation is also straining to commandeer more resources in manpower, materials and equipment than are available? In this environment, prudence and economic sense point to deferral of marginal projects in the interest of lower costs and better quality. But often spending and financing plans only become more elaborate and costly and administrations more determined to carry them out promptly. The tendency of States and localities to ignore the state of the economy in making their commitments to capital spending or borrowing has no simple or obvious explanation. Governments are not motivated -as are corporations by Large fluctuations in cash flow from profits or countercyclical tax concessions. Few sections of the country -2are under such population pressures for new facilities that they cannot defer or alter timing of some projects within the period of several months or even a year. Despite the fact that States and localities have been doing a vastly better job of planning public investment than they did 30 years ago, it probably takes the euphoria of sustained good times to secure final public approval of many shelf-type items. Moreover, in such a booming atmosphere lush scales of private expenditures often have substantial upgrading effects on the quality and type of public facilities. Prosperity unleashes exuberance which unblocks long standing fiscal and institutional restraints. Under these buoyant conditions timing of public works expenditure or borrowing can easily become a secondary consideration. Why wait or why not accelerate? What assurances can be given that there are advantages to altering time schedules in order to save interest or construction costs? Most local financial officials have been advised to borrow when they need the money, not before, and not to attempt to play the bond market. For perfectly obvious reasons I am not prepared to come forward with a bond market forecast today. But I do want to suggest in a realistic actual context how you might view the past behavior of long-term interest rates and some of the factors influencing interest rates for the future. My hope would be to convince you that your attitude toward the interest costs of your governments need not be a know-nothing one and that you can make better judgments on timing than chance will make for you. -3To be as specific as possible, I asked Paul F. McGouldrick of our staff to consider the influences operating on long-term rates over the next three years. His resulting blend of analysis, forecast, and prophecy was prepared under a tight deadline, and hence is essentially speculative and should not be interpreted as more than one expert's insight. Nevertheless, the forecast does provide a convenient frame work within which to consider the points I want to make today. His exercise in crystal ball gazing starts with the following assumptions on the overall state of the nation and the economy: (1) During the next 36 months taken as a whole, GNP in current dollars will grow at an average rate sufficient to keep the labor force close to full employment, but somewhat below the 6.6 per cent per year average growth rate that prevailed between 1960 and 1966. (2) There will be no substantial degree of inflation. While consumer prices will continue to edge higher, the rate of increase will be below the 1966 rate; and the wholesale price index will be close to stable. (3Expenditures directly and indirectly related to national security and the war in South Vietnam will remain at about their current proportion to GNP, or there will be a countervailing expansion of non-military spending if the Vietnam conflict is settled and other military outlays (including a possible anti-missile system) do not offset the consequent saving or a reduction in taxes. His overall conclusion is that it is very probable that the long upward trend in the interest cost of borrowing between 1945 and 1960 will not be resumed during the next three years. Long-term -4- interest rates reached near-historic highs during 1966; and a projection of demands for loanable funds indicates that in the absence of serious inflation, no unusual circumstances exist that would tend to drive interest rates above peaks established over nearly a century of recorded behavior. From 1900 to 1920, these peaks were reached in yields on high-grade railroad bonds which averaged slightly under 6.00 per cent; and the six per cent barrier was breached only for short periods in 1907, 1920, and 1921. The mistakes in Federal Reserve policy in 1920-21 are unlikely to be repeated. During most of the 1920's, both high-grade railroad and high-grade corporate bond yields were somewhat lower than the levels reached during the late summer of last year. Looking still further into the past, only in the 1860's did railroad bond yields break through the highs of last year for broadly comparable securities today (prime corporate bonds); and the picaresque forms which railroad financing took a hundred years ago suggest that quality differences in this comparison are biased in favor of 1861-1870 railroad bonds. In this documentation of the demand for funds in the three years ahead, Mr. McGouldrick expects the catch-up demands for housing, generally forecast for later this year, to slacken off by the end of 1968. The current trends underlying demand for housing thus suggest that, taken as a percentage of GNP, net mortgage borrowing should not rise above 1964 and 1965 levels and, indeed, may even decline slightly. - 5- The current high volume of public issues and private placements of corporate obligations may be expected to decrease during the summer and the fall, as corporations adjust their plant and equipment and inventory spending targets to levels more appropriate to lowered projections of growth in GNP. While a subsequent rebound in corporate debt formation is very likely, as output of goods and services begins to grow at higher rates, this may very well be offset by the satiation of catch-up demand in the mortgage market as 1968 progresses. Household debt formation for the purchase of durables seems unlikely to rise above the high ratios to disposable income that were reached during the early 1960fs; and it is probable that even a sizable absolute increase in the demand for automobiles and other consumer durable goods can and will;co-exist with a slightly lower ratio of net consumer debt formation to personal income than was prevalent two years ago. On balance, then, it appears to him that corporate, mortgage, and consumer debt formation will not rise above 1963-1965 average ratios to GNP. Indeed, the most probable outcome is that these demands for funds will be somewhat lower, as a percentage of GNP, than they were during the early and mid-1960fs. And since these three types of borrowing dominate private markets for borrowed funds, it would take very large Federal debt increases to lift all demands for borrowed funds except those of State and local governments above their average 1961-1965 relationship to GNP. Given this absence of pressure on the demand side, any shortage of funds for State and local borrowers - 6- over the next 36 months would require either a decline in the aggregate propensity to save out of income, or unusual shifts in flows of funds to nonbank financial intermediaries, or a combination of both. On the supply side, he points out that the proportion of savings to disposable income, in the national income and product accounts, which has some year-to-year variability, has shown no significant downward trend since the early 1950's. Indeed, secular stability characterizes this ratio for even longer historical periods, excepting such abnormal eras as the Great Depression and World War II. It would be out of line with historical experience for the ratio of savings to disposable income to fall below the 1961-66 floor of 5.0 per cent, during a period of prosperity* He, therefore, forecasts that the supply of savings by households will be growing at the same percentage rate as will disposable income itself, now that the abnormally low savings rates of early and mid-1966 have been corrected. Business savings are projected to grow at a slower rate than that which prevailed between 1959 and 1966. A substantial part of the huge increase in corporate after-tax profits, during these seven years, was due to a shift of the economy towards a higher level of utilization of non-human capital as well as human resources. Since idle business capital is a source of loss rather than profit, this shift increased after-tax profits by significantly more, proportionately, than it increased GNP and disposable income. But once such a shift has occurred and the economy is around capacity limits, further improvements -7in profits must occur at about the same rate as that by which output increases. Hence, profits are expected to increase, to be sure, but at a markedly lower rate of expansion than that of the previous seven years. Meanwhile, the upward drift of dividend payments may make the rate of increase in retained earnings lower than that of profits. Similar trends are likely in depreciation allowances. Nearly all other factors having to do with both aggregate flows of loanable funds into institutions and capital markets and their distribution among intermediaries argue for an abundant supply of funds for State and local borrowers. For example, demand for municipals by commercial banks may be reasonably expected to remain strong, partly because banks show no signs of losing out to their intermediary competitors (the mutual savings b a n k s > savings and loan associations, and credit unions). To grow significantly, large and small banks must expand their inter mediary role and the instruments and promotion techniques they have used to effect their dramatic gains of the past years will not be abandoned. And since commercial banks are uniquely good customers for State-local tax-exempt securities, because they pay the full corporate income tax on earnings, this would mean a continued and significant growth in one of the principal components of demand for municipals of all maturities. The growing number of higher bracket income recipients will also add to the demand for tax exempts. Rising per capita income before taxes will continue, as before, to increase the pool of individual -8savings seeking direct investment in municipals at a faster rate than that by which aggregate personal income rises. And while brokerage costs and large face amounts of municipals offer obstacles to direct investment by upper-middle-income investors, the recent growth of mutual funds in municipals suggests a more rapid expansion in aggregate demand by individual investors than has occurred over the past 10 years. Over the past six years, there has been a remarkable narrowing of yield differentials among municipals of different quality grades (as shown by Moodys and Standard & Poors ratings). During 1959 and 1960, the differential for Moody's AA and Baa municipals was as high as 80 basis points; but subsequently it narrowed rather persistently. During 1966 and the first three months of 1967, it was between 30 and 55 basis points. Since yields of high-rated municipals did not increase relative to U.S. and corporate bond yields, over the same seven years, this decline in the quality yield spread would appear to signify a greater attractiveness of less than top grade municipals relative to other investment alternatives. When Hr. McGouldrick puts all of the foregoing trends and hypotheses on the U.S. economy together, he comes to the following broad conclusions. Demands for borrowed funds by corporations and households together are not likely to increase at a greater rate than will current dollar GNP. Meanwhile, aggregate savings of households and corporate business are very probably going to increase at the same rate as GNP; and the possibility of a climb in the savings GNP rate should be considered. Finally, a forecast of the changing composition -9of that saving suggests that a gradually increasing, not a falling, fraction will be flowing directly or indirectly into markets for municipal bonds. If this analysis is correct, yields on municipals ought to be declining very moderately over the next 36 months (again, allowing for cyclical fluctuations around the forecast trend), unless State and local demands for funds rise sharply above their trend rate of increase over the postwar period. a number of reasons. And such a sharp increase is very unlikely for One is the rapid expansion of Federal grants-in-aid in connection with the War on Poverty programs. A second has been the unexpected success of State and local governments, considered as a group, in finding new sources of revenue and persuading taxpaying voters to approve increases in rates on existing taxes, particularly since the late 1950's. A third, perhaps more doubtful consideration is the evidence of a certain tapering off of education needs because of the passing of the postwar bulge in elementary and secondary school enrollments. Turning now to the countercyclical timing of construction and financing by State and local governments, what would be the advantages and necessary preconditions for a conscious policy of altering the timing of debt issues with the objective of lowering borrowing costs? The interest yield trend just forecast is not irrelevant to a discussion of the advantages and disadvantages of such a policy. During the long rise in the interest cost of State and local borrowing which culminated during the fifties, municipal treasurers who attempted to wait out periods of tight money were often frustrated by the upward trend -lOunder lying cyclical fluctuations in bond yields. For example, a treasurer might well have concluded, during the last half of 1956, that the current prosperity was due to be succeeded by a business recession and a consequent decline in bond yields. But if he had postponed financing until the first half of 1958, he would probably have paid a higher, not a lower, cost for borrowed funds (the average Moody's triple-A bond yield was 2.69 per cent for July-December 1956 and 2.73 per cent for January-June 1958). In theory, of course, the same upward trend in yields might have acted to encourage accelerated financing during recessions just as it acted to discourage financing postponements during tight money periods. But earlier postwar recessions were relatively short, and the reaction time of State and municipal treasurers was too sluggish to accelerate financing on a large sc^le. Therefore, the net effect of the upward trend in yields prior to 1960 was almost certainly to discourage countercyclical State and local financing. However, the trend in municipal bond yields has levelled off since 1960, and if our forecaster is correct we will have a level or slightly falling trend up to 1970 at least. This should encourage efforts to postpone public works financing in high interest rate periods and to bunch financing in periods when interest rates are relatively low, since correct forecasts will no by an upward drift of yields, of financing practicable? longer be penalized But is such countercyclical timing - 1 1 - During the 14 years since the signing of the Korean War armistice at Panmunjon, our economy has gone through three cycles in interest rates and the availability of credit. Each such cycle culminated in a period of tightness when both short- and long-term interest rates were much higher than before or for some time afterwards. If we measure tightness by such indicators as net borrowed reserves of commercial banks, loan to deposit ratios, and interest rate peaks, and if we read these indicators so as to exclude the effects of underlying trends and irregular fluctuations, we can settle on the following demarcation lines for periods of credit tightness. The first lasted from the early fall of 1956 through October 1957. The second extended from January or February 1959 to January 1960. And the most recent period extended from November or December 1965 to November 1966. All three periods had one characteristic in common which is of particular interest to us. For each lasted for just about the same period of time-- between 12 and 13 months. The point of this exercise in postwar history is to suggest, as strongly as I can, that many State, local, and special authority officials concerned with financing could have saved substantial sums in interest costs for their taxpayers if they had elected to wait out these periods of high interest rates and bond yields, by postponing financing and starts on construction of projects of less immediate essentialness to their communities and regions. Of course, we all know more about events once they have occurred than we do when they are occurring. No one could have said - 1 2 - then, or can say in advance now, during which precise week or month a period of credit tightness will start. Even when such a period has already started, reasonable men can and will differ on the particular month to designate as the start. But during the past 14 years, the great majority of financial economists and analysts have recognized periods of credit tightness for what they were after recognition of a lag of two or three months at the most. Judged by the aforementioned duration of tightness, periods of between 12 and 13 months, this still gave public authorities between 9 and 11 months during which they could have postponed financing. And this also suggests that these authorities will have sufficient time in which to administer such postponements, if a period of credit tightness should conceivably occur during the next three or more years. It has been argued,in another connection, that nearly all types of public construction are for facilities so essential for the communities which they serve, that postponements involve higher social costs than the gain which would accrue from their construction during periods when interest costs are lower. School buildings, for example, must be constructed so as to equate at each time interval the supply of classrooms with the number of pupils, because doubling-up costs far exceed the additional interest cost from building in periods of high interest rates. While this is undoubtedly true, as far as it goes, the argument ignores the presence of a non-negligible category of postponable projects at all times. Examples of such might be: replacement facilities; additions to water, sewer or highway capacity -13for future growth; and marginal projects in the luxury or amenity class. Obviously, timing adjustments require careful planning and exceptions based on special circumstances. For example, rising costs of land often make it inadvisable for many communities to postpone purchases of sites for highways, bridges, and schools; so the best course for them might be to postpone construction but to acquire the sites at once. Another frequently-cited argument against countercyclical variation of public construction is that of rising costs. If, for example, the cost of a project financed by 20-year serial debt issues is forecast to increase»by six per cent next year, the bond yield saving from postponement would have to be a certainty-equivalent of approximately 60 basis points in order for the interest cost saving to offset the construction cost increase. And 60 basis points is a large year-to-year variation in the municipal bond market. This, however, assumes no advance in design, technology and quality over time and that contractors will not figure more closely in a buyers' market. It is correct that postponements involved in a countercyclical capital outlay policy could involve a loss of the type described. ' But, using the same argument, accelerations of projects during periods of relatively low interest rates should involve an offsetting gain. The community would benefit twice, once by the lower interest rates on bond issues, and, secondly, from lower construction costs then than in the future. Since 1960, at least, periods when yields of municipal bonds to investors have been below their 1960-1966 average have actually been longer than periods when - 14- yields were above that seven-year average. If this pattern continues as has been argued, State and local communities will have adequate lead time for accelerating, as well as postponing, construction of less essential roads and highways, buildings, and facilities. Another reason for States and communities to study how to adjust part of their capital spending and borrowing to capital market conditions is that taxpayers bear the full cost of increases in the interest cost of public borrowing while corporations and individuals can pass on a substantial part of increases in their interest payments to the Federal Government (since interest costs are deductible from gross income, for tax purposes). This difference in the impact of higher interest rates also makes corporate borrowing rather insensitive to interest rate changes, so that direct and indirect impacts of tight money tend to fall with peculiar force on supplies of funds for public purposes. While regrettable and even deplorable, this is yet another argument for the proposition that public authorities should pay more attention to capital market conditions in their construction planning. If we focus on borrowing instead of construction spending several additional advantages of countercyclical variations emerge. Even projects of a highly essential nature could, with proper advance planning, be financed more cheaply if public authorities attempt, under favorable market conditions, deliberately first to accumulate stocks of interest-paying liquid assets and then to use these as buffer stocks, so as to permit discontinuities between construction outlays and financing. Borrowing and reinvesting the proceeds in liquid -15assets would appear to be in consonance with the spirit as well as the letter of policies giving public authorities the privilege of borrowing at tax-exempt rates to lenders, as long as there is no primary intent on the part of State and local authorities to profit from the favorable spread between yields of short-term taxable securities (which would constitute their liquid assets) and yields on municipals (which would constitute their liabilities). And the aforementioned spread, which has been moderately favorable during most postwar years, would help to assure authorities that they would not lose by borrowing in order to build up buffer stocks of liquid assets. Such stocks would obviously give communities more latitude in the timing of bond issues than hitherto. And such latitude would be advantageous not only with respect to countercyclical timing but also for variations in timing of bond issues over shorter periods. One thinks, for example, of recent and large "irregular" variations in new issue yields, such as that occurring during the two middle weeks in February when a Western public authority issue of $70 million involved an interest cost about 30 basis points higher than that prevailing on similar obligations issued two weeks earlier. More alertness to such variations would be encouraged, as well as permitted, by accumulations of liquid assets for the purpose just described. -16- To sum up, as a general rule, you should be able to realize significant cost savings in your capital spending programs by adjusting their timing to changing economic and credit conditions. Even though a flexible planning approach of this type might not have had much success during the 1953-60 period of steeply rising bond yields, since 1960--although short-run fluctuations in yields have been large-the underlying trend has been virtually flat. And if Mr. McGouldrick1s forecast is right for the years immediately ahead the trend may actually tip downward. Thus, the potential cost savings to be derived from flexible timing of future capital programs may be quite large. In future periods of relative economic slack, planned outlays on flexible-type programs could be accelerated profitably^producing savings in both construction and financing costs. And more generally, borrowing in periods of relatively low interest rates would add to financial flexibility by creating a cushion of liquid assets. This cushion could be maintained at a positive net interest carry and would be available for the financing of urgent programs in periods when demand pressures on capital markets made it profitable to defer borrowing. To some of you these suggestions may seem a bit gratuitous, like putting my mouth where your money is or substituting a detailed overall view for your accumulated specific experience and responsibility. What you do, however, in response to changes in interest rates and - 17- credit conditions in a general way is my business, as a central banker. For monetary policy, to be effective, changes in the cyclical pattern of aggregate spending are sometimes needed. Thus, in effect, I am urging you to act in your own self-interest, and, in so doing, to help make the impact of public economic policy more effective.
@FedFRASER