Edward Renshaw
Professor of Economics
State University of New York at Albany
In this essay the more extreme values of some changes in economic, financial, political and technical indicators are used to distinguish between positive and negative financial returns for the S&P composite stock price index on a calendar year basis. The most important conclusion to emerge from this approach to stock market forecasting is that optimism pays and so does diversity. The other approaches examined in this essay are not as reliable.
At the end of 1996 there were only two indicators that are summarized in Tables 19.1, 19.2, 19.3 and 19.4 which were pointing in the direction of a positive return for the S&P index for 1997.
While there is not much doubt that the stock market does behave in a cyclical manner the economics profession has not had much luck at developing models that do a good job of forecasting the behavior of stock prices. Trend adjusted changes in stock prices are close to being random in the short run and in a longer run context one has to grapple with the problem of volatile indicators that are not very reliable over a wide range of circumstances.
The first problem encountered at a one year horizon is non linearity. This analyst has never met a stock market indicator that couldn't be improved at the one year horizon by rank ordering changes in the indicator and replacing the more extreme values with one or two dummy variables equal to one and (zero other wise) to separate the more useful extremes from those changes in the indicator that don't do a very good job of discriminating between positive and negative returns (Renshaw, 1993).
Another problem with the unconstrained least squares regression approach to stock market forecasting is that it gives too much weight to "outlying observations" and not enough attention to the finer distinctions or "thresholds" that are able to do the best job of differentiating between good and not so great outcomes.
While there are a number of statistical procedures for identifying threshold effects in variables that are used for forecasting purposes I believe that an investor is well advised to look at the data and exercise some thoughtful discretion in deciding whether an indicator is predicting a good year for the stock market.
In this essay, however, we will constrain our good year predictions to indicator thresholds which would have enabled an investor to have always enjoyed a positive return for the S&P index from 1948-95. In order to take advantage of stock market corrections in the midst of prosperity (as well as crashes that are associated with recessions) it is sometimes desirable to employ a double screen and utilize some secondary extremes to identify good years to have been in the market.
Many of the indicator thresholds discussed in this essay were identified by the author in the 1980s and later published in The Practical Forecasters' Almanac in (1992). During the summer of 1996, however, some additional indicator thresholds were added to this collection in an effort to determine how well various classes of indicators such as monetary, capital and labor market, political and technical indicators can function by themselves. The most important conclusion to emerge from this extension of the good year approach is that narrowly defined classes of indicators cannot be relied upon to achieve perfect discrimination between positive and negative returns and that diversity pays.
Since the justification for some of these indicators and their thresholds may seem a bit "flaky" one can probably be more confident in the approach if numerous indicators are pointing in the direction of a positive return for the S&P index. An extensive overlap of indicator signals, however, can damage an econometric approach by creating an unsolvable problem of "multi-colinearity" that produces crazy regression coefficients that don't do a very good job of discriminating between years with positive and negative returns.
It makes sense, on the other hand, to examine individual indicators from an econometric perspective since confidence in the good year approach can be bolstered to some extent if the current value for the indicator is well above the threshold for outcomes that have been consistently positive in the past and if the threshold has been able to identify a large number of good years in the post 1947 period.
All of the indicators considered in this essay have been able to identity at least six good years to have owned a portfolio similar to the S&P index during the 1948-95 period. When the number of good outcomes is small the indicator is probably being used to "explain" or avoid a rather difficult year in the history of stock market forecasting. For these cases it would seem especially prudent to examine the rationale for "why" the indicator should work.
For the 48 years when three or more of the indicators that are summarized in Tables 19.1, 19.2, 19.3 and 19.4 were pointing in the direction of a good year to own stock the average financial return was equal to 19.5 percent. This can be compared to an average loss of 9.4 percent, for the ten years when none of the indicators were signaling a good time to own stock.
It should be noted, however, that the S&P index is more richly valued now in terms of earnings and dividends received, than was the case from 1942-83, and that good year returns of less than ten percent have been occurring more frequently than was formerly the case.
Evidence in support of a more stable stock market is also apparent in the pattern of bad year returns. There have only been two negative returns since 1977 (1981 and 1990) and both of these losses were rather modest (4.8 and 3.1 percent respectively).
One of the problems with a more stable stock market is that it will leave some of the indicator thresholds that would have worked well in an unstable environment "high and dry" and of little value in predicting what might happen in a more stable environment. In such cases it may make sense to reexamine the data series and consider the possibility of a lower threshold--if it would have functioned better in recent decades.
It should be emphasized that a positive return for the calendar year will not provide investors with complete protection from some nerve racking declines in stock prices which have occurred in the middle of the year. The stock market, however, has outperform money market funds over long periods of time and will no doubt continue to do so in the future.
Positive calendar year returns which have been reduced enough by stock market "corrections" to end the year a bit lower than the alternative return that could have been obtained from Treasury bills have (so far) always been followed by a double digit return for the S&P index in the following year. See those returns identified with an asterisk in the last column of Table 19.2.
Former CEA Chairman Beryl Sprinkel (1964) was among the first economists to publicize a positive relationship between changes in the money supply and changes in stock prices. If one could have accurately predicted what would happen during the year to the real value of the conventional money supply that information would have been of considerable value in helping to identify good years to have been in the stock market.
Since the beginning of World War II there have been 26 years when the December-December growth rate for M1 was at least 1.7 percentage points higher than the growth rate for the all item CPI and in each of these years the financial return for the S&P composite stock price index was positive. (See those cases marked with a double asterisk in column 5 of Table 19.1.) The financial return is defined as the yearly change in the S&P index plus dividends associated with this index during the year expressed as a percent of the value of the index at the end of the preceding year.
When the first differences in this method of adjusting the M1 growth rate for inflation have been equal to four percentage points or more the following year financial returns for the S&P index have also been positive. (See the eleven returns in column 4 of Table 19.1 which are identified with the symbol "MP" and the years identified with an "x" in Table 19.2 under this caption).
Using the data in the first and last columns of Table 19.1 one can show that the financial returns have also been positive after those years when the M1 growth rate increased by three percentage points or more. The presumption is that the Fed would not let the money supply accelerate this much if it believed that inflation was a serious problem that might get out of hand. (These cases are identified with the symbol "M1" in Tables 19.1 and 19.2.
There are two interesting exceptions, however. During 1950, when the Korean War began, the Fed let the growth of M1 accelerate by 4.8 percentage points at a time when the CPI inflation rate increased eight percentage points. Unhappiness with this experience soon led to the celebrated accord between the Fed and the U.S. Treasury in 1951 which freed the Fed from the responsibility of stabilizing the government borrowing rate.
The only other case of a three percentage point acceleration in the M1 growth rate, that wasn't associated with an inflation adjusted increase in the growth rate for M1 of four percentage points or more, occurred at the beginning of another War in 1990 when the Fed was allowing the federal funds rate to decline in an unsuccessful effort to prevent the U.S. economy from slipping into another recession.
The big increases in the M1 growth rate during 1950 and 1990 are worthy of special consideration since they may help to explain the rather impressive financial returns of 23.4 and 30.0 percent for the S&P index during 1951 and 1991.
Increases in the monetary growth rate can also be used in conjunction with other variables to identify good years to have owned common stock. The six years when the monetary growth rate in the first column of Table 13.6 increased and residential building permits in the second column declined on a November-November basis have all been followed by positive returns for the S&P index.
These cases are identified by the symbol "MB" attached to the years 1950, 1951, 1964, 1978, 1986 and 1990 in Table 19.1. When cash and checkable deposits are increasing at an accelerated rate and the demand for mortgages is expected to decline, it is reasonable to suppose that more savings will be available to prop up the stock market.
Investor enthusiasm for equities is also likely to be bolstered if the increase in the monetary growth rate occurs during a stock market correction when many securities can be acquired at bargain prices. For the eight years when the monetary growth rate in the first column of Table 19.1 was up, and the average December value for the S&P composite stock price index was down more than one percent from its preceding average monthly cyclical high, the following year returns for this index were also positive. (These cases are identified with the symbol "MS" in column 1.)
A number of analysts have been perplexed by evidence that common stock returns and changes in the inflation rate have been negatively related to each other over long periods of time. In his (1981) review of the evidence Eugene Fama notes that, "These results are puzzling given the previously accepted wisdom that common stock, representing ownership of the income generated by real assets, should be a hedge against inflation." The inflation fighting policies of the Federal Reserve, however, will probably always make it safer to invest in stocks and bonds when other prices are increasing at a slower rate than when inflation rates are accelerating.
If one could have accurately predicted declines in the CPI, that information would have provided investors with a considerable edge in the stock market. There have been 24 years from the beginning of World War II through 1995 when the inflation rate for the all item CPI in column (2) of Table 19.1 declined by .2 percentage points or more. Except for 1981, when the Fed in its on going war against double digit inflation allowed short term interest rates to rise enough to tip the U.S. economy into an unexpected recession, the financial return on the S&P index in each of these years was positive.
Economists, however, have not had much luck at predicting changes in the inflation rate. There were only four years from 1975-1996 when the President's Council of Economic advisors was able to outperform a no change in the expected growth rate for the fourth quarter to fourth quarter inflation rate for the implicit price deflator for GNP or GDP (Table 12.6).
Investors, on the other hand, have exhibited a propensity to bid up stock prices after tighter monetary policy has been successful at keeping the inflation rate from accelerating. There have been ten years since the beginning of World War II when the M1 growth rate in column (1) of Table 19.1 declined by two percentage points or more and the CPI inflation rate in column (2) either declined or experienced an increase of only .1 percentage points.
All of these cases, which are identified by the symbol "CPM", were followed by a positive return for the S&P index. This is the only good year indicator that this analyst has discovered which can explain the sparkling return of 35.7 percent during the war time year of 1945 when more than 12 million U.S. citizens were on active duty with the armed services.
There is another way in which investors have responded favorably to what might be regarded as very tight monetary policy. Since the great depression of the 1930s the U.S. stock market has always been followed by a positive financial return after any year in which the Fed's M2 or Friedman and Schwartz's M4 increased on a December-December basis by four percentage points or less.
The slowdown in the growth of M2 during the first half of the 1990s was related in part to very low rates of return on checkable deposits, CD's and money market funds which encouraged a maturing baby boom to pump record amounts of savings into long term stock and bond funds.
Martin Zweig, whose stock market newsletter outperformed 24 other newsletters during the 1980s, has developed a complicated set of economic and financial indicators to predict what will happen in the stock market. In his book, Winning on Wall Street, Zweig's main advice to investors is "Don't Fight the Fed."
There have been ten years since the end of World War II when the Fed eased monetary policy enough to allow the December-December yield on new issues of three month Treasury bills to decline by 21 percent or more (Table 17.4). These years are identified in Tables 19.1 and 19.2 with the symbol "TB" and were all followed by a positive return for the S&P index.
Section 2A of the Federal Reserve Act as amended, requires the Board of Governors "to promote effectively the goals of maximum employment, stable prices and moderate long-term interest rates". Moreover, in carrying out monetary policy the Fed is to take "account of past and prospective developments in employment, unemployment, production, investment, real income, productivity, international trade and payments and prices."
In keeping with its responsibility to keep a close eye on the economy the Fed has developed its own highly publicized index of economic activity: industrial production. This index has an impressive record of enabling one to take advantage of stock market crashes in the vicinity of recessionary peaks in business activity.
The nine years since 1947 when industrial production declined by one percent or more on a December-December basis were all followed by double digit returns with an average value of 30.5 percent. These years are identified in Table 19.2 with the caption "IP."
It should be emphasized, however, that stock market indicators that would have worked fairly well on a calendar year basis some times break down and are less reliable on an interim basis. There have been several occasions when industrial production experienced a cumulative decline of one percent or more in the midst of prosperity, or in the early stages of a recession, when an investor would have had to wait more than a year to have been able to liquidate an index fund tracking the S&P composite at a profit.
The financial returns associated with the S&P index have been better, on the average, during the two years preceding a presidential election than in post election years. To find a negative return for this index in a presidential election year one has to go all the way back to the turmoil of the great depression of the 1930s and the dark days preceding U.S. entry into World War II.
Since 1941 all of the negative returns for the S&P stock price index have occurred either in post election years or in the following year. See the more favorable returns identified with the symbol "POL" in Tables 19.1 and 19.2.
Since the outcome of presidential elections is sensitive to what has been happening in the economy during the two years before an election (Trahan and Renshaw 1990) and since the Fed's freedom to manage short term interest rates and manipulate the money supply could be reduced by Congress, it is reasonable to suppose that the Board of Governors would prefer to keep a low profile before a presidential election.
The two years culminating in a presidential election, in any event, does allow one to identify six years with positive financial returns (1947, 1956, 1963, 1964, 1967 and 1980) that are not easy to explain on the basis of what was happening to money, inflation and short term interest rates in the preceding year.
While none of the individual indicators examined in Tables 19.1 and 19.2 are very reliable predictors of what will happen in the stock market they can collectively identify all but one of the positive return years for the S&P index from 1943-95 without any false signals.
The one omission is 1970, when the S&P index declined almost 25 percent before bottoming out on May 26 and clawing its way upward to an anemic gain of .1 percent for the year as a whole. If the dividends received during 1970 were reinvested at the end of the year the financial return would have only been 3.5 percent during a year when the yield on new issues of three month Treasury bills was averaging 6.46 percent.
There have been five other years (1960, 1978, 1984, 1987 and 1994) when a positive return for the S&P index was not quite sufficient to offset the return that could have been obtained on Treasury bills. Investors who remained in the stock market during these years, however, would not have had to wait very long to be ahead of a policy of keeping one's savings in, say, a money market fund. After each of these "sub-par" years the S&P index rebounded enough to provide patient investors with a double digit return.
The success of monetary and political indicators at identifying good times to have been in the stock market is a little surprising considering all of the political upheavals and inflationary shocks which have occurred since the great depression of the 1930s.
The usefulness of statistical indicators based on changes in the more conventional money supply, M1, is also a bit puzzling considering the tremendous amount of financial innovation which has occurred during the last two decades. These innovations have forced the Board of Governors of the Federal Reserve to "down grade" the behavior of both M1 and M2 and rely more heavily on other types of economic and financial indicators before announcing changes in the federal funds rate.
While portfolio managers are probably well advised to examine other types of good year indicators before purchasing equities, more than fifty years of turbulent history would suggest that it might be a mistake to ignore the behavior of monetary indicators altogether. In Table 19.3 the emphasis will be on capital and labor market indicators.
The growth of corporate earnings has been the key to long run success in the stock market. Record earnings for the S&P index, however, have not done a very good job of protecting investors from near term bear markets. The problem with record earnings is that they are usually "anticipated" and not a very reliable guide to future price appreciation. In 1928, 1929, 1965, 1968, 1972, 1973 and 1976 the earnings associated with the S&P index increased 24.3, 16.7, 14.1, 8.1, 12.6, 27.1 and 24.5 percent respectively and in the following years the financial returns were -7.9, -23.9 -10.0, - 8.3, -14.5, -26.0 and -7.2 percent.
There is a possibility, however, that investors are now more impressed with record earnings than was formerly the case. None of the nine records established from 1977-96 have (so far) been followed by a negative financial return for the S&P index (Table 18.4).
Since the beginning of World War II, though, it has been safer to have invested in the stock market after a yearly decline in the earnings associated with the S&P 500 of six percent or more than after an increase of six percent or more. See those "x" mark the spot years in summary Table 19.3 that are captioned with an "E".
Since the imposition of wage and price controls during World War II and the Korean War most of the larger declines in corporate earnings have been associated with economic recessions. The only exception in recent decades is 1985 when a slowing economy, declining oil prices and a prolonged period of very rapid appreciation in the value of the US dollar was making it very difficult for manufacturing companies located in the US to sell their output at a profit.
It should be appreciated that stock prices are a leading economic indicator and that earnings come closer to being a coincident indicator. By the time recessionary declines in corporate earnings are publicized much of the gain to be expected from easier monetary policy and declining interest rates may have already occurred. In the post 1947 period several employment indicators would have provided investors with a more timely way to take advantage of economic recessions.
Price/Earnings ratios are among the most closely watched stock market indicators. The stock market has performed better on the average, over long periods of time, in a low P/E ratio environment than in a high P/E environment. Since the stock market crash of 1929 there have been nine years when a portfolio similar to the S&P index could have been acquired at an end of the year P/E ratio under 9.00. All of these years were followed by a positive financial return.
It is doubtful, however, whether investors will soon have an opportunity to acquire the S&P index at such a low value in relation to earnings. The volatility of earnings and fluctuations in the way investors have valued earnings over time have made the vaunted P/E ratio, by itself, an unreliable indicator of what will happen in the stock market (Essay 18).
There were five occasions from 1927-89 when the end of the quarter P/E ratio for the S&P composite stock price index increased to a value in excess of 20.50. After each of these events an investor could have sold a portfolio similar to this index and repurchased it during the next bear market at a discount of at least 16 percent (Renshaw 1990, Table 1).
An investor following this type of trading signal would have exited the market at the end of September, 1991 when the S&P closed at 387.86 and remained out of the market until at least the end of the first quarter of 1994 when the P/E had fallen to 19.6 and the index closed at 445.77. He or she would have been spared some of the trauma associated with the Fed's preemptive strike against accelerating inflation in February 1994 but would have incurred some transaction costs and would have had to repurchase the S&P index at a premium of about 15 percent. If the stock market "correction" of 1994 had scared the investor enough to refrain from reentering the market he or she would have missed out on a great time to own equities in 1995 and 1996.
Dividend growth rates also have the property of being adversely affected by economic recessions. Since the beginning of World War II investors in a portfolio similar to the S&P index would have always enjoyed a positive financial return after any year when the dividend growth rate for this index experienced a decline of 3.2 percentage points or more. See those years captioned with a "D" in Table 19.3.
Since most of the large declines in the dividend growth rate that are not associated with wars occurred in the vicinity of economic recessions it is fairly easy to find other indicators that are as useful at identifying good years to have been in the stock market.
There is a more intriguing way that dividends can be used to help identify good years to have owned common stock, however. Since 1941 the S&P index has always registered a positive return after any year when its average dividend yield increased by 24 basis points or more. See the column labeled "DIV" in Table 19.3 for the occurrence of such cases. An increase in the dividend yield of this magnitude can be caused by an increase in dividends and/or a decline in stock prices.
Since large increases in the dividend yield are more likely to be the result of a pronounced bear market, rather than a rapid increase in corporate dividends, this type of signal can be of some value in helping investors to take advantage of stock market corrections in the midst of prosperity as well as the crashes associated with economic recessions. Since the prolonged recession of 1973-75 the financial returns for the S&P index have been positive after any increase in its average dividend yield.
In Essay 4 it was noted that when the actual growth of real fixed investment has been at least 2.1 percentage points greater than the expected increase that was predicted by an "AR" accelerator model optimism on the part of business enterprises and developers has tended to spread to the stock market and generate positive following year financial returns for the S&P index.
This indicator was persistently optimistic from 1992 through 1996 and may help to explain why the Fed induced stock market correction of 1994 was much more modest than the corresponding bond market correction. It may also help to explain why the stock market was able to appreciate in a fairly impressive way during the first half of 1996 when the bond market was in disarray.
Evidence in support of a non random pattern to stock prices is easy to overlook because it often involves non linear relationships and first or second differences in a statistical indicator. The validity of this conclusion can be illustrated to some extent by examining large changes in the average unemployment rate.
There have been seven years since the end of World War II when the unemployment rate increased by 1.4 percentage points or more on a year-year basis and each of these recessionary years was followed by a positive return for the S&P index. There have been nine years since 1940 when the unemployment rate declined by one percentage point or more and each of these rather vigorous years of recovery from a recession inspired enough investor confidence to be followed by a positive return for the S&P index. See those indicator signals captioned with a "UN" in Table 19.4.
A more stable economy, however, has reduced the frequency of such signals. Investors who were trying to take advantage of the recession which began in July 1990 would have been better off to have focused their attention on December values for the civilian unemployment rate that were up at least one percentage point from its yearly low. See those cases in Table 19.3 that are captioned with a "U".
Persons who are trying to take advantage of economic recessions should also keep a close eye on declines in payroll employment and increases in initial claims for unemployment insurance.
There were eight years from 1948-95 when payroll employment or "EMP" in Table 19.3 declined on a Nov.-Nov. basis. The average following year return for the S&P composite stock price index was 22.9 percent.
An even higher average following year return is associated with the 12 cases when the initial unemployment claims or "UC" in Table 19.3 increased by 13.5 percent or more on a Nov.- Nov. basis. All but the 1995 case for this indicator are associated with economic recessions.
Payroll employment and initial unemployment claims are also of some value in helping one to take advantage of stock market corrections in the midst of prosperity.
There have been eleven years since 1946 when there was no increase in the S&P stock price index and payroll employment declined or increased by less than 3.5 percent. All of these years, which are captioned with the symbol "SE" in Table 19.3, were followed by a positive return for the S&P index.
The "SU" symbol represents years when the November-November percentage change in initial unemployment claims was negative and the average December value of the S&P index was at least 1.7 percent less than its preceding monthly average cyclical high. All of these years were followed by a positive financial return.
One reason why it is easier to identify good years to have owned common stocks than bad years is related to the fact that bull markets and business expansions have more staying power, on the average, than bear markets and economic recessions.
Since the beginning of World War II there has only been one bear market that was sufficiently prolonged to produce negative returns for the S&P index in two consecutive calendar years--the bear market of 1973-74. During the same period of time we have always had back-to-back returns of a positive nature after each transition from a negative to a positive financial return.
The symbol "UR" is used to identify the financial returns following an upside reversal in Table 19.3. The average return for these years has been equal to 16.1 percent.
This measure of bull market persistence and the ten capital and labor oriented stock market indicators in Table 19.3 would have enabled an investor to have avoided the ten financial return loss years from 1948-95 and to have benefitted from all but three of the positive return years. The three positive returns that are not identified with this collection of signals all fall in the mediocre range of from .3 for the recessionary year of 1960 to 5.7 percent for the boom and bust year of 1987. An investor utilizing these signals would have also avoided a return of 5.6 percent for 1956, which was followed by a recessionary loss of 10.4 percent in 1957.
History would suggest, in any event, that an investor should be rather cautious if there are no capital or labor market indicators pointing in the direction of a good year to own common stock. In Table 19.4 the emphasis will be on cyclical indicators.
In a now classic article titled, "Can Stock Market Forecasters Forecast?" Alfred Cowles (1933) concluded: "It is doubtful." Another long term follower of efforts to predict what might happen in the stock market, Nobel Laureate Paul Samuelson, has voiced similar skepticism.
In his (1989) review of efforts to refute the random walk hypothesis Samuelson concluded: "Broadly speaking, the case for efficient markets is a bit stronger in 1989 than it was in 1974, or in 1953 when Holbrook Working and Maurice Kendall were hypothesizing that stock and commodity price changes are pretty much a random walk (or a white-noise martingale). Out of the thousands of published and unpublished statistical testings of various forms of the hypothesis, a few dozen representing a minuscule percentage have isolated profitable exceptions to the theory."
This perception of the stock market, however, is in the process of changing. Fortune (1991) has summarized a number of studies which have cast considerable doubt on the random walk hypothesis. The presumption that the stock market behaves in the manner of a random walk, moreover, is not very credible when one considers its behavior in relation to the business cycle
The best years to have owned common stock, on the average, have been years containing a recessionary trough in economic activity as defined by the National Bureau of Economic Research. Since the stock market crash of 1929 the financial returns associated with the S&P composite stock price index have averaged 31.6 percent during trough years (Table 3.2). This can be compared to a more risky average return of only 9.75 percent during the years which have immediately followed a recessionary trough.
In the post World War II period economic recessions have been of such short duration that by the time a majority of forecasters are convinced that the economy is in a recession, it is probably time to begin looking for bargains in the stock market. At this point in the business cycle one doesn't need leading indicators to identify good years to be in the market. Most of the coincident indicators identified by the National Bureau of Economic Research will also be of some forecasting value.
In the post 1947 period there have been ten occasions when the Conference Board's composite index of four coincident indicators failed to increase on a November-November basis and each of these years was followed by a positive return for the S&P index. See those returns in the next to the last column of Table 19.4 that are identified with an asterisk.
Since the great depression of the 1930s years of recovery from a negative financial return for the S&P index have so far always been followed by a second positive return. See those returns identified with a hatch mark in Table 19.4.
History suggests that one can probably be more confident about the good year approach to forecasting if there are a large number of indicators pointing in the direction of a great time to own common stock than if there are only a small number of good year signals. This point can be verified to some extend by also considering some over lapping indicators which were first published in The Practical Forecasters' Almanac in 1992 and by tabulating the total number of good year signals covered in Tables 19.2, 19.3 and 19.4 which are summarized in the last column of Table 19.4.
When the absolute value of the current year's financial return for the S&P index is subtracted from the annual high-low ratio, expressed as a percentage point range, one is left with an intriguing measure of stock market volatility that would have been especially helpful at enabling an investor to take advantage of major declines in stock prices. When this measure of "residual volatility" has exceeded 16 percentage points, the financial returns for the S&P index in the following two years have always been positive since the beginning of World War II. See those initial and following year designations under the caption of "RV" in Table 19.4.
Recessionary crashes and stock market corrections in the midst of prosperity often lead to a reduction in trading activity. Bad years for brokers, however, are often an indication of good times to acquire more stock. The seven years when there was a ten percent or more decline in the December-December volume of shares traded on the New York Stock Exchange, captioned by "VOL" in Table 19.4, were all followed by a positive return for the S&P index.
While investors have usually been better off to have purchased stock after a decline in a statistical series with leading economic properties, there have been some exceptions. There have been nine years since 1947 when the Fed allowed interest rates to decline enough to encourage builders to increase housing starts by 25 percent or more on a November-November basis. Each of these housing start bubbles listed in Table 19.4 under an "HS" designation was followed by a positive return for the S&P index.
There were nine years from 1942 to 1982 when the Dec.-Dec. growth rate for the all item CPI was at least 1.4 percentage points less than the year- to-year increase in the CPI inflation rate. The following year financial returns for the S&P index for these cases ranged from a low of only 5.4 percent in 1948 to a high of 31.0 percent in 1955 and have an average value of 20.4 percent.
Most of these signals, however, occurred when the year-year growth rate for the CPI was in the 6.1 to 14.4 percent range. It is doubtful whether the Fed will allow the CPI inflation rate to again get that high in the remainder of this century.
The thirty or more indicators summarized in Tables 19.1, 19.2, 19.3 and 19.4 can explain all of the positive returns for the S&P index since 1947 without any false signals. When ten or more of these indicators were pointing in the direction of a good year to own common stock the following year returns for the S&P index have averaged 22.2 percent and have always exceeded the average discount rate on new issues of three month Treasury bills.
Except for 1978, which benefitted from easier monetary policy and an exceptionally large number of corrections in the midst of prosperity signals following the prolonged decline in stock prices from September 21, 1976 to March 6, 1978, all of the cases involving ten or more good year signals are associated with economic recessions.
Years containing a recessionary peak in business activity have usually been poor years to own common stock. Since the S&P index was first computed on a daily basis in 1928, there have only been four positive returns for the recessionary peaks occurring in: 1945, 1948, 1960 and 1980, when the financial returns were 35.7, 5.4, .3 and 31.5 percent respectively.
The S&P stock price index was down a bit for the year in both 1948 and 1960. The positive financial returns for these two years were entirely dependent upon dividends received during the year. The other two cases were associated with short lived recessions where the peaks and troughs occurred in the same year.
At the end of 1944 real GDP expressed in 1987 dollars was more than twice as great as it was in 1929 and the S&P index could have still been acquired at a discount of more than 58 percent from its 1929 high.
The only other recessionary peak year with an impressive financial return is 1980, when the peak occurred in January and the recessionary trough occurred in July of the same year. At the end of 1979 the P/E ratio for the S&P index was at the lowest level since the stock market crash of 1973-74.
Some mad speculation in energy producing companies after the sky rocketing oil prices, which followed the political turmoil associated with the overthrow of the Shah in Iran, also helped to insure a prompt recovery from the 17.1 percent recessionary slump in the S&P index from February 13 to March 27, 1980.
Iranian oil production, which had averaged 5.2 million barrels per day in 1978, fell to 3.2 million in 1979 and 1.7 million in 1980 before bottoming out at a low of only 1.4 million barrels per day in 1981.
Throughout this collection of essays we have tried to arranged economic and financial statistics in a manner that lets the data speak for itself. In academia and even in the real world, however, there is a propensity to value complexity for its own sake. In his book on Long- Range Forecasting (Wiley, 1978), J. Scott Armstrong notes that:
"Simple methods are as accurate as complex ones. Although this is contrary to the belief of experts in econometrics, none of 11 empirical studies examined showed a significant gain in accuracy from using complex methods. In fact, the data suggested the opposite.If you are going to play the stock market game, however, it makes sense to at least have a score card and make some effort to understand how this market has behaved in relation to economic and financial indicators that are publicized by the news media and fairly easy to up-date."Throughout Long-Range Forecasting I have emphasized the value of simple methods: they are cheaper, easier to implement and often more accurate than complex methods. But the big money goes into complexity. Why?
"The rain dance has something for everyone. The dancer gets paid. The clients get to watch a good dance. The decision maker gets to shift the problem onto someone else in a socially acceptable way. (Who can blame him? He hired the best dancer in the business)."
In his book on Stocks for the Long Run (Chapter 17) Jeremy Siegel has noted that September is the worst month to have owned common stock on the average and the only month with a negative return. "Maybe the poor returns in late fall have nothing directly to do with economics but are related to the approach of winter and the depressing effect of the shorter days". Siegel goes on to suggest that the poor returns may be "the result of investors liquidating stocks (or holding off buying new stocks) to pay for their summer vacations."
An even more chilling depressant is the way we finance education. Some educators are paid on a nine or ten month basis and don't received their first pay check until about the end of September. School taxes are often due in September and some parents may be forced to liquidate stock to pay for tuition and the expense of sending their children to private schools, colleges and universities. September is also one of those months when a successful investor has to pay his or her estimated income taxes.
Financial innovations which have made it easier and less costly for some investors to switch retirement funds between stock and money market funds may have made it more rational for investors who are not burdened with educational expenses in September to delay their own stock purchases or move out of equities in the hope of being able to repurchase them at a lower price after the "correction" which has so often occurred in September.
In late August of 1995, Siegel's findings with regard to a bad September were widely publicized by Steve Sakson of the Associated Press. The S&P composite stock price index, instead of declining, continued to recover from the July 19th low of 550.98 and beginning on September 5th proceeded to achieve eight consecutive new historic closing highs. This was the first time that many new closing highs had been achieved in an eight day trading period since July of 1964.
On September 3, 1996 it was reported in the Wall Street Journal that since 1915 September and October are the only two months to provide, on average, negative returns for the Dow Jones Industrial Average. Both the Dow and the S&P 500 then proceeded to drift upward and record some more new highs.
One of the more important lessons to be learned from this type of experience is: never trust an average. It can conceal a lot of period to period variability. In Table 19.5 we show the monthly percentage changes in The S&P composite stock price index since it was extended back to 1928.
From 1928-96 there were 39 years when the S&P index declined during September and only 29 years when it increased. Before jumping ship or delaying stock purchases, however, it makes sense for one to look for relationships that might enable an investor to do a better job of distinguishing between good and bad Septembers.
In this essay the emphasis has been on economic and financial indicators that would have enabled an investor to have differentiated between years with positive and negative financial returns. For the 20 years from 1928-95 when the financial returns for the S&P index were negative, this index experienced an average loss of 4.635 percent during September. For the 48 years when the financial return for the year was positive, the S&P index averaged a slight gain of .15 percent during September.
The stock market is more likely to suffer prompt corrections in a turbulent new high bull market than in a market which achieves numerous new highs without a lot of day-to-day volatility. One measure of volatility that can easily be determined from the data in Table 19.5 is the severity of the monthly downside corrections for the S&P index during the calendar year prior to September.
For the eleven years from 1928-95 when the worst monthly correction prior to September was a decline of 2.4 percentage points or less (1944, 1955, 1958, 1959, 1964, 1972, 1976, 1985, 1987, 1992 and 1995) the average following September percentage change has been positive.
Another factor that individual investors ought to keep in mind in appraising the September outlook is the possibility that money managers have become more forward looking in their behavior. For the 16 years when the S&P index increased in July and then declined in August (perhaps in anticipation of a bad September) the stock market has advanced more often than not in September and since 1959 has consistently increased in September.
Consecutive declines in the S&P index in June, July and August (which occurred in 1946, 1966, 1974, 1981 and 1990), on the other hand, were all followed by declines of more than five percent in September. The last three of these cases are associated with economic recessions.
The possibility that a very bad summer for the stock market might help to tip the US economy into an economic recession leads this analyst to conclude that the Federal Reserve ought to be encouraged to buy and sell stock index futures to help stabilize this market, reduce the risk of economic recessions and perhaps offset, to some extent, a saving and investment imbalance in September when people go back to work and school.
The invention of stock index futures makes it possible for investors with a portfolio similar to the S&P index to hedge their portfolio from a bad September. Let us assume that it is the end of August and that the S&P index has experienced a monthly increase of more than one percent. Should an investor try to lock in profits by selling a futures contract? See Table 19.6.
The stock market crashes of 1929 and 1987 have given October a bad reputation. One of the more interesting questions to be asked in connection with Table 19.5 is whether there can be a bad October without a disappointing September? For those cases where there was an increase in the S&P index of .1 percent or more in September the worst following decline in October was only 3.0 percent in 1955--a loss that was more than compensated for by an increase in stock prices of 7.5 percent in November.
For the eight years from 1928-95 when September gains were equal to 4.0 percent or more, the average following October change for the S&P index was a gain equal to .56 percent--in spite of four losses amounting to -1.5, - 1.9, - .1 and - .5 percent in 1939, 1954, 1973 and 1995.
From 1928-94 there were actually fewer monthly declines for the S&P index in October than in February, May, June, September and November. The safest months to have owned a portfolio similar to the S&P index have been December and January.
February used to be a bad month for stocks. Since the stock market correction of 1984, however, there have only been two declines in the S&P index during February. February's superior performance in recent years may be related in part to the much publicized January effect.
January and December are probably the most interesting months to study the stock market. By the mid 1970s statisticians were beginning to conclude that there might be some seasonality to the monthly rates of return from holding common stock due primarily to large January returns and by the early 1980s it was discovered that most of the exceptional returns in this month were associated with small firms.
While much has been written about the January size effect it is less well appreciated that this month often sets a tone for the market that will carry forward to the end of the year. From 1947-96 there were 27 years when the S&P composite stock price index increased by more than one percent in January and in every one of these years the financial return for the entire year has (so far) been positive. (See those cases marked with an asterisk in the first column of Table 19.5)
The financial return for the entire year was also in excess of the January gain in all but two of these years. The first exception was 1987 when the S&P index, after the largest January gain in its history (13.2 percent), increased another 22.9 percent before getting involved in the most spectacular new-high crash in the history of the NYSE. The other exception is 1994 when the Fed launched a preemptive strike against the possibility of accelerating inflation beginning in February and continued to raise short term interest rates through February of 1995.
The identification of relatively safe years to have invested in the stock market can be improved somewhat by combining the gains for both January and the preceding December. For the 29 cases from 1947-95 where the combined gain was in excess of three percent for these two months the full year returns for the S&P index are also positive.
The unresolved question is whether this success story is an accident or whether there is something special about January that enables investors to more accurately anticipate the behavior of stock prices in the remainder of the year.
It has been noted that January is the start of the tax year for investors and the beginning of the tax and accounting years for most firms. If this month gets off to a good start some funds that were taken out of the market in connection with tax-loss selling in the preceding year may be reinvested.
Preliminary and in many cases final announcements of previous calendar (fiscal) year's sales and earnings are often made in January. Some analysts have suggested that the extra information that is available at this time of the year from both government and the private sector may enable investors to more accurately forecast the future.
Except for the short lived recession from January 1980 to July of that year--when the Fed was experimenting with some very high interest rates in a determined effort to end the specter of double digit inflation--the US economy has never experienced a peak in business in any year since 1945 when the January gain in stock prices was over one percent. Let us hope that 1997 will not turnout to be another exception.
The stock market, though over exuberant at times, is one of those games where records are made to be broken. Ten or more new historic highs for the S&P 500 in the first quarter of a calendar year have (so far) always been followed by at least six new highs in the second quarter. Ten or more new highs in the second quarter have also been followed by at least one more new high in the third or fourth quarter. See Table 19.7 which shows all of the new highs for the S&P composite index which were recorded on a quarterly basis from 1928 to March 1997.
This table also provides one with some perspective on the severity of those stock market crashes which have followed new high bull markets. So far there has never been a major crash that began in the second quarter of a calendar year. Five of the eleven worst new high declines, however, did begin in the third quarters of 1929, 1956, 1959, 1987 and 1990.
Nervous participants in the stock market, however, can perhaps take some comfort in the fact that investors have not had to wait very long for a full recovery from recent crashes. During the 52 year period from 1928-79 there were only 16 years with one or more new highs for the S&P index. In the 16 year period since 1979 there have only been three years (1981, 1983 and 1988) without any new highs.
If we measure the duration of a major bull market on the basis of new highs separated by a cumulative decline of 11 percent or more, the bull markets in Table 19.7 range in duration from only two quarters for the 1980 bull market--which was terminated by Paul Volcker's successful effort to end the specter of double digit inflation by raising short term interest rates to unprecedented levels--to a rather unprecedented 23 quarters, and still counting, for the 1991-97 bull market--which (as of February 28, 1997) had not yet suffered a "correction" of more than 8.9 percent.
Trying to out guess the market is an interesting and rather challenging pastime. It is made difficult by many minor corrections in the midst of more prolonged bull markets. From 1954 to May 24, 1996 there were 133 dips of from one to 10.6 percent in the S&P composite index in the midst of major bull markets. They make it hard to identify the end of a bull market until stock prices have declined enough to make many equities a bargain. Most investors, therefore, are probably well advised to savor each new high, when it does occur, but follow the strategy of Peter Lynch and invest for the long run. Stocks, after all, are the only type of financial asset that can reasonably be expected to set many more "appreciation" records with the passage of time.
Nerve racking declines in the prices of some technology stocks have revived memories of 1987 and led some analysts to suggest that this may be the prelude to a major crash. It is hard to find convincing evidence in support of such a conclusion, however.
Since the invention of stock index futures in the early 1980s, which make it easier for hedgers and speculators to "short" an over valued market, stock market declines have not lasted very long. Since the prolonged recession of 1981-82 none of the cumulative new high declines for the S&P composite stock price index amounting to 7.5 percent or more have lasted more than 199 trading days.
The spectacular one day loss of more than 20 percent for the S&P index on October 19, 1987 occurred on the 38th trading session after the August 25 peak. During the summer of 1990, after Iraq's invasion of Kuwait, it only took 28 trading days for the S&P index to lose 16.8 percent of its value.
While no one can predict what will happen to the stock market with much confidence there are a number of historical relationships which suggest that the S&P index will probably end this year on a positive note.
At the end of 1996 there were at least two indicators (summarized in Tables 19.1, 19.2, 19.3 and 19.4) which were pointing in the direction of a positive financial return for the S&P index during 1997.
One factor that ought to be kept in mind during stock market corrections is the behavior of business enterprises. When real fixed investment has been increasing much more rapidly than might be expected on the basis of the simple "accelerator" model which was discussed in Essay 4, optimism on the part of developers and business enterprises has tended to spill over to the financial sector and produce positive following year financial returns for the S&P index.
To find a time of greater investment enthusiasm on the part of business enterprises one has to go all the way back to the end of World War II. Since the recessionary trough year of 1991 the real value of business investment in new equipment has increased by more than 50 percent at a time when our gross domestic product was increasing by less than 14 percent.
The big upward surge in equipment spending has helped to more than double the earnings associated with the S&P index. In this type of environment, which can't last forever, it is reasonable to expect enthusiasm for technology and other types of emerging "growth stocks" to get a bit out of hand and be followed by some major price corrections for those companies whose sales and earnings don't live up to rosy expectations.
When the stock market decline that began in February 1997, is viewed from this perspective it can be considered a "healthy development" and not something to get overly alarmed about. The US stock market has outperformed "money market funds" over long periods of time and will no doubt continue to do so in the future.
History, in any event, provides strong support for the notion that investors would have been better off, on the average, to have purchased an index fund tracking the S&P index after days when it declined by one percent or more than after days when it increased by at least one percent (Table 17.7).
In most years a policy of buying after a bad day on Wall Street would have also been superior to a dollar cost averaging strategy of investing equal amounts of money in the market at the end of every trading day.
Since the stock market crash of 1987 it has usually taken longer for the S&P index to recover to another new historic high, after a downside correction of 7.5 percent or more, than it took to reach its low point for the correction. See the last two columns of Table 19.8
The data in Table 19.7 would suggest, however, that there is a good chance that the S&P Index will register at least one more new high before 1997 is over. Since the S&P Index was extended back to 1928 on a daily basis cumulative new high declines of three percent or more beginning in March, April, May or June have (so far) always been followed by at least one more new historic high in either the third or the fourth quarter of the same calendar year. See Tables 19.7 and 20.1.
While it is easy to devise systems for moving in and out the stock market that would have provided trading profits (before taxes and transaction costs) superior to a policy of buying and holding a representative portfolio, this analyst has not had much luck at devising crash or bad year indicators that performed well in an out of sample period.
The best way to profit from nonrandomness, I believe, is to buy stock on margin after there has been a large decline in stock prices and numerous indicators of both a cyclical and value oriented nature are pointing in the direction of lower interest rates and a rapid recovery of stock prices (Renshaw 1995).
There is one bad year indicator that was first publicized by (Renshaw in 1990) that ought to be kept in mind, however, during a prolonged bull market that might turnout to end in a "bursting bubble". From 1954-90 a policy of selling the S&P index after two double-digit financial returns in a row and repurchasing it during the next bear market would have provided investors with a discount ranging from a low of 7.5 which would have been obtainable during the crash of 1987 to a high of 47.2 percent during the more horrendous crash of 1973-74. See Table 19.9.
The unanswered question as of March 1, 1997 is whether inexperienced investors and money managers have taken control of the US stock market and are setting the stage for a crash that will end the most prolonged new high bull market in the history of the S&P composite stock price index.
Cowles, Alfred (1933). "Can Stock Market Forecasters Forecast?" Econometrica, (July) 309-24.
Fama, E. (1981). "Stock Returns, Real Activity, Inflation and Money," American Economic Review, 545-65.
Fortune, Peter (1991). "Stock Market Efficiency: An Autopsy?" New England Economic Review, March/April, 17-40.
Renshaw, Edward (1990). "Some Evidence in Support of Stock Market Bubbles," Financial Analysts Journal, March/April, 71-73.
-----, (1992). The Practical Forecasters' Almanac(Irwin Professional Publishing).
-----, (1993). "Modeling the Stock Market for Forecasting Purposes," Journal of Portfolio Management, 20(Fall), 76-81.
-----, (1995). "There's No Big Picture, or, That Is the Big Picture," The Journal of Investing, Summer, 56-61.
Trahan, Emery and Edward Renshaw (1990). "Presidential Elections and the Federal Reserve's Interest Rate Reaction Function," Journal of Policy Modeling, 12(1), 29-34.
Samuelson, Paul (1989). "The Judgment of Economic Science on Rational Portfolio Management: Indexing, Timing, and Long Horizon Effects," The Journal of Portfolio Management, (Fall), 4-12.
Siegel, Jeremy (1994). Stocks for the Long Run(Irwin Professional Publishing).
Sprinkel, B. (1964). Money and Stock Prices(Richard Irwin).
Year Dec.-Dec. Growth Rates Real First Financial Return
M1 CPI M1 Diff. S&P Composite
(1) (2) (3)n (4)n (5)
1942 29.3M1 9.0 20.3 15.9MP 19.2**
1943 27.0CPM 3.0* 24.0 3.7 25.7** POL
1944 13.4CPM 2.3* 11.1 -12.9 9.3** POL
1945 12.9 2.2 10.7 - .4 35.7**
1946 5.2 18.1 -12.9 -23.6 - 7.8
1947 4.4 8.8* - 4.4 8.5MP 5.5 POL
1948 - 1.4CPM 3.0* - 4.4x .0 5.4 POL
1949 - .3 - 2.1* 1.8x 6.2MP 17.8**
1950MB 4.5M1 5.9 - 1.4x - 3.2 30.5
1951MB 5.6 6.0 - .4 1.0 23.4 POL
1952 3.8 .8* 3.0 3.4 17.7** POL
1953TB 1.1CPM .7 .4 - 2.6 - 1.2
1954TB 2.7 - .7* 3.4 3.0 51.2**
1955 2.2 .4 1.8 - 1.6 31.0** POL
1956 1.3 3.0 - 1.7 - 3.5 6.4 POL
1957 - .7CPM 2.9 - 3.6 - 1.9 -10.4
1958 3.8M1 1.8* 2.0 5.6MP 42.4**
1959 .6CPM 1.7 - 1.1x - 3.1 11.8 POL
1960TB .5 1.4* - .9 .2 .3 POL
1961 3.2 .7* 2.5 3.4 26.6**
1962 1.8 1.3 .5 - 2.0 - 8.8
1963 3.7 1.6 2.1 1.6 22.5** POL
1964MB 4.6MS 1.0* 3.6 1.5 16.3** POL
1965 4.7 1.9 2.8 - .8 12.3**
1966 2.4 3.5 - 1.1 - 3.9 -10.0
1967 6.6M1 3.0* 3.6 4.7MP 23.7** POL
1968 7.7 4.7 3.0 - .6 10.8** POL
1969 3.3 6.2 - 2.9 - 5.9 - 8.3
1970TB 5.1 5.6* - .5 2.4 3.5
1971 6.5MS 3.3* 3.2 3.7 14.1** POL
1972 9.2 3.4 5.8 2.6 18.7** POL
1973 5.5 8.7 - 3.2 - 9.0 -14.5
1974 4.4 12.3 - 7.9 - 4.7 -26.0
1975TB 4.8MS 6.9* - 2.1 5.8MP 36.9 POL
1976 6.6 4.9* 1.7 3.8 23.6** POL
1977 8.1MS 6.7 1.4 - .3 - 7.2
1978MB 8.2MS 9.0 - .8 - 2.2 6.4
1979 6.8 13.3 - 6.5 - 5.7 18.4 POL
1980 6.8 12.5* - 5.7 .8 31.5 POL
1981TB 6.8 8.9* - 2.1 3.6 - 4.8
1982TB 8.7 3.8* 4.9 7.0MP 20.4**
1983 9.8MS 3.8 6.0 1.1 22.3** POL
1984 5.9CPM 3.9 2.0 - 4.0 6.0** POL
1985 12.3M1 3.8 8.5 6.5MP 31.1**
1986MB,TB 16.9M1 1.1* 15.8 7.3MP 18.5**
1987 3.5 4.4 - .9x -16.7 5.7 POL
1988 4.9MS 4.4 .5 1.4 16.3 POL
1989 .9 4.6 - 3.7 - 4.2 31.2
Year Dec.-Dec. Growth Rates Real First Financial Return
M1 CPI M1 Diff. S&P Composite
(1) (2) (3)n (4)n (5)
1990MB 4.0MS,M1 6.1 - 2.1x 1.6 - 3.1
1991TB 8.6M1 3.1* 5.5x 7.6MP 30.0** POL
1992TB 14.2M1 2.9* 11.3x 5.8MP 7.4** POL
1993 10.2CPM 2.7* 7.5x - 3.8 9.4**
1994 1.8CPM 2.7 - .9x - 8.4 1.3
1995 - 2.1CPM 2.5* - 4.6 - 3.7 37.1 POL
1996 - 4.3 3.3 - 7.6 - 3.0 22.3 POL
Footnotes for Table 19.1
(3)n. Column (1) minus column (2).
(4)n. First differences in column (3).
CPM identifies years when the M1 growth rate declined by two percentage points or more on a December-December basis and the all item CPI inflation rate declined or increased by .1 percentage point or less. All of these years have been followed by a positive financial return for the S&P index.
M1 identifies years when the growth of M1 in column (1) increased by three percentage points or more. All of these years have been followed by a positive return for the S&P index.
MB identifies years when the growth rate for M1 in column (1) increased and there was a November-November decline in residential building permits. All of these years have been followed by a positive return for the S&P index.
MP identifies years when the first differences in the inflation adjusted growth rate for M1 in column (4) were equal to 4.0 percentage points or more. All of the financial returns following these years have been positive.
MS identifies years when the growth rate for M1 increased and the average value for the S&P index in December was down one percent or more from its preceding monthly average cyclical high. All of these corrections in the stock market have been followed by a positive return for the S&P index.
POL identifies presidential election years and preelection years. The financial returns for the S&P index have been positive in these years.
TB identifies years when the December-December yield on new issues of 3- month Treasury bills was allowed to decline by 21 percent or more. All of these years have been followed by a positive return for the S&P index.
x identifies years when the Fed's broader monetary aggregate M2 and Friedman and Schwart's M4 increased by four percent or less. All of these years have been followed by a positive return for the S&P index.
* identifies years when the December-December growth rate for the CPI declined by .2 percentage points or more. Except for 1981, when the Fed allowed short term interest rates to rise enough to tip the US economy into a recession, the financial returns for these years have all been positive.
** identifies cases where the inflation adjusted growth rate for M1 in column (3) was equal to 1.7 percentage points or more. All of the financial returns in these years have been positive.
Following Year
Year POL M1 MP MB MS M2 TB CPM IP S&P Return
1947 x x 5.4T
1948 x x 17.8
1949 x x x 30.5
1950 x x x x 23.4
1951 x x 17.7
1952--------------------------------------------------- - 1.2
1953 x x x 51.2
1954 x x 31.0
1955 x 6.4
1956--------------------------------------------------- -10.4
1957 x x 42.4
1958 x x x 11.8
1959 x x x .3*
1960 x x 26.6
1961--------------------------------------------------- - 8.8
1962 x 22.5
1963 x 16.3
1964 x x 12.3
1965--------------------------------------------------- -10.0
1966 x 23.7
1967 x x x 10.8
1968--------------------------------------------------- - 8.3
1969 3.5*
1970 x x x 14.1
1971 x x 18.7
1972--------------------------------------------------- -14.5
1973--------------------------------------------------- -26.0
1974 x x x 36.9
1975 x x x 23.6
1976--------------------------------------------------- - 7.2
1977 x 6.4*
1978 x x x 18.4
1979 x 31.5
1980--------------------------------------------------- - 4.8
1981 x x 20.4
1982 x x x x 22.3
1983 x x 6.0*
1984 x 31.1
1985 x x 18.5
1986 x x x x x 5.7*
1987 x x 16.3
1988 x 31.2
1989--------------------------------------------------- - 3.1
1990 x x x x x x 30.0
1991 x x x x x 7.4
1992 x x x x 9.9
1993 x x 1.3*
1994 x x x 37.1
1995 x x 22.3
1996--------------------------------------------------- ?
Footnotes for Table 19.2.
POL indicates that a presidential election year will follow either the current year or the next year.
M1 identifies years when the growth rate for M1 increased by 3.0 percentage points or more.
MP identifies years when the first differences in the inflation adjusted growth rates for M1 in column (4) of Table 19.1 were equal to four percentage points or more.
MB identifies years when there was an increase in the M1 growth rate and a decline in residential building permits.
MS identifies years when the growth rate for M1 increased and the average value for the S&P index in December was down one percent or more from its preceding monthly average cyclical high.
M2 identifies years when the growth rate for the Fed's M2 or Friedman and Schwartz's M4 was equal to four percent or less.
TB identifies years when the December-December yield on new issues of 3- month Treasury bills was allowed to decline by 21 percent or more.
CPM identifies years when the M1 growth rate declined by two percentage points or more on a December-December basis and the all item CPI inflation rate declined or increased by .1 percentage points or less.
IP identifies years when industrial production decline by one percent or more on a December-December basis.
* identifies positive financial returns for the S&P index that were a bit lower than the alternative return that could have been obtained from Treasury bills. Each of these positive, but sub-par returns were followed by a double digit return for the S&P index.
Following Year
Year E P/E D DIV AR UC EMP U SE SU UR S&P Return
1947 x x x x 5.4
1948 x x x x x x 17.8
1949 x x x x x 30.5
1950 x 23.4
1951 x x 17.7
1952---------------------------------------------------------- - 1.2
1953 x x x 51.2
1954 x x x 31.0
1955 6.4
1956---------------------------------------------------------- -10.4
1957 x x x x x x 42.4
1958 x x x x 11.8
1959 .3
1960 x x x 26.6
1961---------------------------------------------------------- - 8.8
1962 x x x 22.5
1963 x 16.3
1964 x 12.3
1965---------------------------------------------------------- -10.0
1966 x x x 23.7
1967 x x 10.8
1968---------------------------------------------------------- - 8.3
1969 x x 3.5
1970 x x x x x x x x 14.1
1971 x x 18.7
1972---------------------------------------------------------- -14.5
1973---------------------------------------------------------- -26.0
1974 x x x x x 36.9
1975 x x x x x 23.6
1976---------------------------------------------------------- - 7.2
1977 x x x x 6.4
1978 x x x x x x 18.4
1979 x x x 31.5
1980---------------------------------------------------------- - 4.8
1981 x x x x 20.4
1982 x x x x x x x x 22.3
1983 x 6.0
1984 x x 31.1
1985 x 18.5
1986 5.7
1987 x 16.3
1988 x x 31.2
1989---------------------------------------------------------- - 3.1
1990 x x x x x 30.0
1991 x x x x 7.4
1992 x 9.9
1993 x 1.3
1994 x x x 37.1
1995 x x 22.3
1996 x ?
Footnotes for Table 19.3
E identifies a year when the earnings associated with the S&P index declined on a year-year basis by six percent or more.
P/E identifies year end price earnings ratios for the S&P index that were less than 9.0.
D represents years when the dividend growth rate for the S&P 500 declined at least 3.2 percentage points
DIV represents years when the average dividend yield for the S&P index increased by at least 24 basis points.
AR represents years when the difference between the actual and the predicted growth rate for real fixed investment in Table 1.32 of The Practical Forecasters' Almanac was equal to 2.1 percentage points or more.
UC represents years when initial unemployment claims increased by 13.5 percent or more on a Nov.-Nov. basis.
EMP represents years when there was a Nov.-Nov. decline in payroll employment. The average following year financial return in these years was 22.9 percent.
U represents cases where the December unemployment rate was up at least one percentage point from its yearly low.
SE represents years when there was no increase in the S&P composite stock price index and the November-November percentage change in payroll employment was less than 3.5 percent.
SU represents years when the November-November percentage change in initial unemployment claims was negative and the average December value of the S&P composite stock price index was at least 1.7 percent less than its preceding monthly average cyclical high.
UR represents a first year of recovery from a negative return year.
Following Year Total Good
Year HLR RV VOL HS UN CP S&P Return Year Signals
1947 x x 5.4# 9
1948 x x 17.8 10
1949 f x x 30.5* 11
1950 x 23.4 6
1951 x x x 17.7 7
1952--------------------------------------- - 1.2 0
1953 x x 51.2 8
1954 f x x x 31.0*# 10
1955 x x x 6.4 4
1956--------------------------------------- -10.4 0
1957 42.4* 8
1958 x x 11.8# 10
1959 x x .3 5
1960 26.6 5
1961--------------------------------------- - 8.8 0
1962 x x 22.5 6
1963 x f 16.3# 5
1964 12.3 3
1965--------------------------------------- -10.0 0
1966 x x 23.7 6
1967 x f x 10.8# 9
1968--------------------------------------- - 8.3 0
1969 x 3.5 3
1970 x x x 14.1*# 15
1971 x f x 18.7 7
1972--------------------------------------- -14.5 0
1973--------------------------------------- -26.0 0
1974 x x 36.9* 10
1975 x f x x x 23.6*# 14
1976--------------------------------------- - 7.2 0
1977 x 6.4 6
1978 x x 18.4# 12
1979 x f 31.5 6
1980--------------------------------------- - 4.8 0
1981 x x x 20.4* 9
1982 x x x x 22.3*# 17
1983 x f x 6.0 6
1984 x 31.1 4
1985 18.5 3
1986 x 5.7 6
1987 x 16.3 4
1988 x f x 31.2 6
1989--------------------------------------- - 3.1 0
1990 x x 30.0* 13
1991 f 7.4*# 11
1992 x 9.9 6
1993 1.3 3
1994--------------------------------------- 37.1 6
1995--------------------------------------- 22.3 4
1996 x ? 2
Footnotes for Table 19.4
HLR represents years when there was a downside reversal for the yearly high- low ratio for the S&P composite stock price index.
RV represents differences of 16 percentage points or more between the high- low range for the S&P index expressed in percentage points minus the current financial return for the S&P index. The letter "f" identifies a first follow on year.
VOL represents years with December-to-December declines in the volume of shares traded on the NYSE of ten percent or more.
HS represents years when housing starts increased 25 percent or more on a November-November basis.
UN represents years when the average civilian unemployment rate either increased by 1.4 percentage points or more or declined by one percentage point or more.
CPI represents years when the December-December growth rate for the all item CPI was at least 1.4 percentage points less than the year-year increase in the CPI inflation rate.
* represents years when there was no increase in the Conference Board's composite index of four coincident indicators on a November-November basis.
# identifies financial returns following a transition from negative to positive financial returns for the S&P index.
Yr. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 28 -1.1 -1.8 10.8 3.2 1.3 -4.0 1.3 7.4 2.4 1.5 12.0 .3 29 5.7* -.6 -.2 1.6 -4.3 11.2 4.6 9.8 -4.9#-19.9 -13.4 2.5 30 6.2* 2.2 8.0 -1.0 -1.6 -16.5 3.7 .8 -13.0# -8.9 -2.2 -7.4 31 4.9* 11.4 -6.9 -9.6 -13.7 13.9 -7.4 .9 -29.9# 8.4 -9.8 -14.5 32 -2.8 5.1 -11.8 -20.2 -23.3 -.9 37.7 37.5 -3.7#-13.9 -5.9 5.2 33 .7 -18.4 3.4 42.2 15.9 13.2 -8.8 11.5 -11.4 -8.9 10.3 2.2 34 10.6* -3.7 -.1 -2.7 -8.1 2.1 -11.5 5.4 -.5# -3.2 8.3 -.4 35 -4.2 -4.0 -3.1 9.6 3.2 6.8 8.3 2.2 2.4 7.5 3.9 3.7 36 6.6* 1.7 2.5 -7.7 4.6 3.1 6.8 .9 .1 7.5 .4 -.6 37 3.8* 1.5 -.9 -8.3 -1.0 -5.3 10.3 -5.5 -14.2#-10.2 -10.1 -5.0 38 1.3* 6.1 -25.0 14.1 -4.4 24.7 7.3 -2.7 1.5 7.6 -3.3 3.8 39 -6.9 3.3 -13.5 -.5 6.2 -6.4 10.9 -7.1 16.5# -1.5 -4.9 2.4 40 -3.5 .7 1.0 -.5 -24.0 7.7 3.1 2.6 .9# 3.9 -4.2 -.3 41 -4.8 -1.5 .4 -6.5 .4 5.3 5.5 -.9 -1.0# -6.9 -4.2 -4.5 42 1.4* -2.5 -6.8 -4.4 6.4 1.8 3.1 .7 2.7 6.4 -1.4 5.2 43 7.2* 5.1 5.3 .1 4.5 2.0 -5.4 1.0 2.4 -1.3 -7.6 5.9 44 1.5* -.3 1.7 -1.2 4.0 5.1 -2.1 .9 -.3 .0 .4 3.5 45 1.4* 6.2 -4.6 8.8 1.1 -.3 -2.0 5.8 4.2 3.0 3.2 1.0 46 7.0* -6.9 4.6 3.8 2.2 -3.9 -2.6 -7.3 -10.2# -.8 -1.1 4.3 47 2.4* -1.5 -1.7 -3.9 -.9 5.3 3.6 -2.8 -1.4 2.1 -2.9 2.1 48 -4.0 -4.7 7.7 2.7 7.8 .3 -5.3 .8 -3.0 6.8 -10.8 3.1 49 .1 -3.9 3.0 -2.1 -3.7 -.2 6.2 1.2 2.4 3.0 .1 4.4 50 1.7* 1.0 .4 4.5 3.9 -5.8 .8 3.3 5.6 .4 -.1 4.6 51 6.1* .6 -1.8 4.8 -4.1 -2.6 6.9 3.9 -.1 -1.4 -.3 3.9 52 1.6* -3.6 4.8 -4.3 2.3 4.6 1.8 -1.5 -2.0 -.1 4.6 3.5 53 -.7 -1.8 -2.4 -2.6 -.3 -1.6 2.5 -5.8 .1# 5.1 .9 .2 54 5.1* .3 3.0 4.9 3.3 .1 5.7 -3.4 8.3 -1.9 8.1 5.1 55 1.8* .4 -.5 3.8 -.1 8.2 6.1 -.8 1.1 -3.0 7.5 -.1 56 -3.6 3.5 6.9 -.2 -6.6 3.9 5.2 -3.7 -4.5 .5 -1.1 3.5 57 -4.2 -3.3 2.0 3.7 3.7 -.1 1.1 -5.6 -6.2# -3.2 1.6 -4.1 58 4.3* -2.1 3.1 3.2 1.5 2.6 4.3 1.2 4.8 2.5 2.2 5.2 59 .4 -.0 .1 3.9 1.9 -.4 3.5 -1.5 -4.6 1.1 1.3 2.8 60 -7.1 .9 -1.4 -1.8 2.7 2.0 -2.5 2.6 -6.0 -.2 4.0 4.6 61 6.3* 2.7 2.6 .4 1.9 -2.9 3.3 2.0 -2.0 2.8 3.9 .3 62 -3.8 1.6 -.6 -6.2 -8.6 -8.2 6.4 1.5 -4.8# .4 10.2 1.3 63 4.9* -2.9 3.5 4.9 1.4 -2.0 -.3 4.9 -1.1 3.2 -1.1 2.4 64 2.7* 1.0 1.5 .6 1.1 1.6 1.8 -1.6 2.9 .8 -.5 .4 65 3.3* -.1 -1.5 3.4 -.8 -4.9 1.3 2.3 3.2 2.7 -.9 .9 66 .5 -1.8 -2.2 2.1 -5.4 -1.6 -1.3 -7.8 -.7# 4.8 .3 -.1 67 7.8* .2 3.9 4.2 -5.2 1.8 4.5 -1.2 3.3 -2.9 .1 2.6 68 -4.4 -3.1 .9 8.2 1.1 .9 -1.8 1.1 3.9 .7 4.8 -4.2 69 -.8 -4.7 3.4 2.1 -.2 -5.6 -6.0 4.0 -2.5# 4.4 -3.5 -1.9
Yr. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 70 -7.6 5.3 .1 -9.0 -6.1 -5.0 7.3 4.4 3.3 -1.1 4.7 5.7 71 4.0* .9 3.7 3.6 -4.2 .1 -4.1 3.6 -.7 -4.2 -.3 8.6 72 1.8* 2.5 .6 .4 1.7 -2.2 .2 3.4 -.5 .9 4.6 1.2 73 -1.7 -3.7 -.1 -4.1 -1.9 -.7 3.8 -3.7 4.0# -.1 -11.4 1.7 74 -1.0 -.4 -2.3 -3.9 -3.4 -1.5 -7.8 -9.0 -11.9# 16.3 -5.3 -2.0 75 12.3* 6.0 2.2 4.7 4.4 4.4 -6.8 -2.1 -3.5 6.2 2.5 -1.2 76 11.8* -1.1 3.1 -1.1 -1.4 4.1 -.8 -.5 2.3 -2.2 -.8 5.2 77 -5.1 -2.2 -1.4 .0 -2.4 4.5 -1.6 -2.1 -.2# -4.3 2.7 .3 78 -6.2 -2.5 2.5 8.5 .5 -1.8 5.4 2.6 -.7 -9.2 1.7 1.5 79 4.0* -3.7 5.5 .2 -2.6 3.9 .9 5.3 .0 -6.9 4.3 1.7 80 5.8* -.4 -10.2 4.1 4.7 2.7 6.5 .6 2.5 1.6 10.2 -3.4 81 -4.6 1.3 3.6 -2.3 -.2 -1.0 -.2 -6.2 -5.4# 4.9 3.7 -3.0 82 -1.8 -6.1 -1.0 4.0 -3.9 -2.0 -2.3 11.6 .8 11.0 3.6 1.5 83 3.3* 1.9 3.3 7.5 -1.2 3.5 -3.3 1.1 1.0 -1.5 1.7 -.9 84 -.9 -3.9 1.3 .5 -5.9 1.7 -1.6 10.6 -.3 -.0 -1.5 2.2 85 7.4* .9 -.3 -.5 5.4 1.2 -.5 -1.2 -3.5 4.3 6.5 4.2 86 .5 7.1 5.3 -1.4 5.0 1.4 -5.9 7.1 -8.5 5.5 2.1 -2.8 87 13.2* 3.7 2.6 -1.1 .6 4.8 4.8 3.5 -2.4 -21.8 -8.5 7.3 88 4.0* 4.2 -3.3 .9 .3 4.3 -.5 -3.9 4.0 2.6 -1.9 1.5 89 7.1* -2.9 2.1 5.0 3.5 -.8 8.8 1.6 -.7 -2.5 1.7 2.1 90 -6.9 .9 2.4 -2.7 9.2 -.9 -.5 -9.4 -5.1# -.7 6.0 2.5 91 4.2* 6.7 2.2 .0 3.9 -4.8 4.5 2.0 -1.9 1.2 -4.4 11.2 92 -2.0 1.0 -2.2 2.8 .1 -1.7 3.9 -2.4 .9 .2 3.0 1.0 93 .7 1.0 1.9 -2.5 2.3 .1 -.5 3.4 -1.0 1.9 -1.3 1.0 94 3.3* -3.0 -4.6 1.2 1.2 -2.7 3.1 3.8 -2.7 2.1 -4.0 1.2 95 2.4* 3.6 2.7 2.8 3.6 2.1 3.2 -.0 4.0 -.5 4.1 1.7 96 3.3* .7 .8 1.3 2.3 .2 -4.6 1.9 5.4 2.6 7.3 -2.2 97 6.1*
*Identifies cases where the S&P index increased one percent or more during January. Since 1946 the financial return for the entire year has (so far) always been positive for these cases.
#Identifies September changes which are associated with negative annual financial returns for the S&P index.
Source of basic data: Standard and Poor's Security Price Index Record.
Yr. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
82 1.3 -3.5 10.7 -.0 12.8 3.2 1.8
83 3.9 1.6 1.0 8.1 -1.2 4.2 -3.7 .6 1.4 -1.3 .8 -.9
84 -1.1 -5.0 1.1 .3 -6.4 1.0 -1.6 9.9 -.1 -.8 -2.7 1.9
85 6.1 1.4 -1.4 -1.5 4.9 .3 -1.4 -2.0 -3.7 3.4 7.3 3.7
86 .4 6.3 5.3 -2.8 5.4 1.0 -6.6 7.1 -9.3 6.0 1.5 -2.8
87 13.4 3.5 2.1 -.6 .1 4.1 5.5 3.4 -2.4 -20.8 -10.7 5.9
88. 4.1 4.0 -3.6 .8 .7 4.0 - .7 -4.2 3.8 2.3 -2.5 1.6
89 6.6 -2.9 1.5 4.6 3.1 -1.4 8.4 1.2 -1.0 -2.8 1.4 1.2
90 -7.2 .7 1.4 -2.7 8.9 -1.2 -1.1 -10.0 -6.0 .1 5.6 1.2
91 4.2 6.4 1.7 -.2 3.7 -5.2 4.4 1.8 -2.0 .7 -4.5 10.8
92 -2.3 .9 -2.3 2.6 -.0 -1.6 3.4 -2.3 .9 -.0 3.3 1.3
93 .1 1.3 1.9 -3.1 2.7 .0 -.7 3.4 -1.0 1.9 -1.3 .8
94 3.1 -3.2 -4.4 .8 1.4 -3.0 3.1 3.3 -2.8 2.0 -3.9 .9
95 2.2 3.4 2.4 2.4 3.6 1.3 2.9 .0 3.7 -.7 4.1 .9
96 3.1 .1 .1 .6 1.9 .6 -5.1 1.4 5.2 2.6 6.9 -2.7
97 5.8
Arithmetic Average Percentage Change Since the Beginning of Futures
2.8 1.3 .5 .7 2.1 .4 .2 1.6 - .9 .3 .6 1.7
Source of basic information: The Wall Street Journal.
Number of New Historic Highs per Quarter
Year ---------------------------------------- % Change S&P Index
First Second Third Fourth Last Peak to Trough
(1) (2) (3) (4) (5)
1928 12* 9 13 24
1929 10* 9 29***----------------- -86.2
1954 5 21
1955 6 23** 16 1
1956 7 0 7***----------------- -21.6
1958 3 21
1959 11* 8 8-------------------- -13.9
1961 17* 9 9 19---------- -28.0
1963 5 6
1964 27* 14** 10 11
1965 10* 15** 1 11
1966 9---------------------------------------- -22.2
1967 3 11 0
1968 0 11** 5 17---------- -36.1
1972 3 8 5 16
1973 3---------------------------------------- -48.2
1980 14*** 10---------- -27.1
1982 2
1983 12* 16** 0 1---------- -14.4
1985 13* 11** 6 13
1986 16* 9 5 1
1987 26* 7 14***----------------- -33.5
1989 8 5
1990 0 5 1-------------------- -19.9
1991 5 6 4 7
1992 4 0 5 9
1993 6 0 6 4
1994 5---------------------------------------- - 8.9
1995 14* 30** 17 16
1996 10* 6 4 19
1997 14* ?
Footnotes for Table 19.7
*Cases where there were ten or more new highs in the first quarter of a calendar year. These cases have always been followed by six or more new highs in the second quarter.
**Identifies ten or more second quarter new highs. None of these cases have been followed by a major stock market crash beginning in the third quarter of the year in question.
***Cases of an acceleration of seven or more new highs in the third quarter during a prolonged bull market. These cases were soon followed by stock market crashes of twenty percent or more.
Closing Values Number of Trading Days
-----Date of------ ---S&P Index--- Percent from Peak Trough to
Peak Trough Peak Trough Decline to Trough Recovery
10/10/83 7/24/84 172.65 147.82 14.38 199 125
7/17/85 9/25/85 195.65 180.66 7.66 49 33
7/02/86 7/15/86 252.70 233.77 7.53 8 30*
9/04/86 9/29/86 253.83 229.91 9.42 17 45*
4/06/87 5/20/87 301.95 278.21 7.86 31 17
8/25/87 12/04/87 336.77 223.92 33.51 71 414*
10/09/89 1/30/90 359.80 322.98 10.23 78 82*
7/16/90 10/11/90 368.95 295.46 19.92 62 86*
2/02/94 4/04/94 482.00 438.92 8.94 41 219*
5/24/96 7/24/96 678.51 626.65 7.64 41 36
*Cases where the recovery from the trough to a new historic high for the S&P index took longer than the correction from peak to trough.
Financial S&P Index Values Percentage Decline
Return ------------------------- After Two Double
Year (Percent) High Low Close Digit Returns
(1) (2) (3) (4) (5)n
1953 - 1.2 26.66 22.71 24.81 ---
1954 51.2 35.98 24.80 35.98 ---
1955 31.0 46.41 34.58 45.48* ---
1956 6.4 49.74 43.11 46.67 ---
1957 -10.4 49.13 38.98L 39.99 -14.3
1958 42.4 55.21 40.33 55.21 ---
1959 11.8 60.71 53.58 59.89* ---
1960 .3 60.39 52.30L 58.11 -12.7
1961 26.6 72.64 57.57 71.55 ---
1962 - 8.8 71.13 52.32 63.10 ---
1963 22.3 75.02 62.69 75.02 ---
1964 16.3 86.28 75.43 84.75* ---
1965 12.3 92.63 81.60 92.43 ---
1966 -10.0 94.06 73.20L 80.33 -13.6
1967 23.7 97.59 80.38 96.47 ---
1968 10.8 108.37 87.72 103.86* ---
1969 - 8.3 106.16 89.20 92.06 ---
1970 3.5 93.46 69.29L 92.15 -33.3
1971 14.1 104.77 90.16 102.09 ---
1972 18.7 119.12 101.87 118.05* ---
1973 -14.5 120.24 92.16 97.55 ---
1974 -26.0 99.80 62.28L 68.56 -47.2
1975 36.9 95.61 70.04 90.19 ---
1976 23.6 107.63 90.90 107.46* ---
1977 - 7.2 107.00 90.71 95.10 ---
1978 6.4 106.99 86.90L 96.11 -19.1
1979 18.4 111.27 96.13 107.94 ---
1980 31.5 140.52 98.22 135.76* ---
1981 - 4.8 138.12 112.77 122.55 ---
1982 20.4 143.02 102.42L 140.64 -24.6
1983 22.3 172.65 138.34 164.93* ---
1984 6.0 170.41 147.82L 167.24 -10.4
1985 31.1 212.02 163.68 211.28 ---
1986 18.5 254.00 203.49 242.17* ---
1987 5.7 336.77 223.92L 247.08 - 7.5
1988 16.6 283.66 242.63 277.72 ---
1989 31.2 359.80 275.31 353.40* ---
1990 - 3.1 368.95 295.46L 330.22 -16.4
1991 30.0 417.09 311.46 417.09 ---
1992 7.4 441.28 394.50 435.71 ---
1993 9.9 470.54 429.05 466.45 ---
1994 1.3 482.00 438.92 459.27 ---
1995 37.1 621.69 459.11 615.93 ---
1996 22.3 757.03 598.48 740.74* ?
Footnotes for Table 19.9
*Closing value for second year of double digit return in column (1).
L equals subsequent bear market low.
(5)n. Percentage decline from the double digit close in column (4) to the subsequent bear market low.
Source of basic data: Standard and Poor's Security Price Index Record.
Go on to Essay 20:
Return to the Introduction