THIS IS A STUDY of comparative rates of inflation in various groups of countries over the postwar period, based on a detailed examination of increases in cost of living indices.1 The existence or absence of inflationary pressures is not, of course, necessarily indicated by marked upward movements of price indices, such as the cost of living index. Reserve and exchange rate changes may absorb some of the inflationary pressures in some countries during particular periods. Also, price movements may be repressed in different degrees in different countries at different times. Therefore, price changes alone may not indicate fully the degree of inflationary pressure. Nevertheless, relative price increases provide at least an a priori indication of whether inflationary pressures have been more prevalent in one group of countries than another.
Cost of living indices and indices of wholesale prices are the only price series that are commonly available. Wholesale price indices are considered less satisfactory as measures of domestic prices because they “are frequently heavily weighted with the prices of imports and exports and tend to be measures of the domestic equivalents of international prices rather than measures of domestic prices.” 2 This is not, of course, to say that cost of living indices are free of defects.3 Nevertheless, they are used here, as they are used by others, because they are the least unsatisfactory of the available measures of domestic price increases.
Aspects of the rates of inflation that are examined in this study include the averages of rates of inflation, the year-to-year variability of these rates, and the trends in them. These aspects are examined not only for the whole of the period 1949–65 but also for the three subperiods into which it is divided: 1949–53, 1954–59, and 1960–65. In examining rates of inflation, primary emphasis is placed on the comparison of price movements in different groups of countries. Data for the entire period covered were available for only 53 countries. These countries were classified into three groups: (1) industrial—13 countries; (2) more developed primary producing, herein referred to as other developed—10 countries; and (3) less developed primary producing—30 countries. However, the price changes were so much larger in 6 of the 53 countries—Argentina, Bolivia, Brazil, Chile, Korea, and Uruguay—that these countries seem clearly to form a group by themselves.4 All 6 of these countries are in the less developed group; their inclusion in that group under the classification described above is bound to affect very sharply the picture of the price behavior of less developed countries in general. It was decided, therefore, to present data not only for all the less developed countries as a group but also for the group of less developed countries excluding these 6 high-inflation countries.
I. Average Rates of Inflation
In this section, attention is directed primarily to the average rates of inflation that were recorded in each country and in the different groups of countries during the period under review. The annual rates of inflation for the period 1949–65 are presented for each country in Table 9 in the Appendix. The average rates of inflation for the different groups of countries are presented in Table 1 for the entire period as well as for the three subperiods.
|Other developed countries||4.85||6.06||4.79||3.91|
|Less developed countries||9.89||11.99||10.00||8.03|
|Less developed countries excluding high-inflation countries||3.63||5.04||2.73||3.38|
An interesting point shown in Table 1 is that, during the period 1949–65, none of the average rates for the different groups of countries was greater than 10 per cent. Even when the subperiods are examined, it is found that an average rate of inflation greater than 10 per cent was recorded only once—for the more inclusive less developed countries group during 1949–53.
For 1949–65 as a whole, the average rate is higher for the more inclusive less developed group than for the other developed group, and the rate for the latter is higher than that for the industrial group of countries. In fact, the average rate recorded by the more inclusive less developed group is at least twice the rates recorded by the other two groups. However, when the 6 high-inflation countries are excluded, the average rate of inflation for the less developed group of countries during the period compares favorably with those of the other two groups. Indsed, this less inclusive group of less developed countries had an average rate of inflation slightly lower than that of the industrial countries and much lower than that of the other developed group.
The observed relationships between the average rates in the three groups of countries during the subperiods are the same as those observed for the whole period. In all three subperiods, the more inclusive less developed group of countries had higher average rates than those of the other two groups. However, excluding the 6 high-inflation countries, the less developed group had average rates of inflation that were lowest in the first and third subperiods. In the second subperiod the average rate of 2.73 per cent was slightly higher than the 2.35 per cent for the industrial countries, but it was much lower than the 4.79 per cent rate recorded by the other developed group.
Of the three subperiods, 1949–53 witnessed the highest average rates of inflation in all the groups of countries. In the industrial group, the average rate dropped sharply during the second subperiod but increased in the third subperiod. In the other developed group, the average rates fell in each succeeding subperiod. While this also was the case in the more inclusive less developed group, the behavior of the average rates in the less inclusive less developed group was similar to that observed in the industrial group.
For convenience of more detailed discussion, the rates of inflation have been arbitrarily divided into four groups: (1) rates of price increase not greater than 2 per cent are described as Relative Price Stability; (2) rates of inflation over 2 per cent but not more than 5 per cent, Mild Inflation; (3) rates between 5 per cent and 10 per cent, Moderate Inflation; and (4) rates over 10 per cent, High Inflation.
Table 2 presents the proportional distribution of the various groups of countries according to these ranges of inflation. In only the first subperiod did more than half of the countries (28) have rates of inflation that could be called either Moderate or High, i.e., rates greater than 5 per cent. For 1949–65 as a whole, 20 countries (38 per cent)—and in both the second and third subperiods, 14 countries (26 per cent)—had such rates of inflation.
|A. Not more than 2 per cent||B. Over 2 per cent but not more than 5 per cent||A + B. Not more than 5 per cent|
|Less developed excluding high-inflation countries||33||33||50||42||42||25||29||38||75||58||79||80|
|C. Over 5 per cent but not more than 10 per cent||D. Over 10 per cent||C + D. Over 5 per cent|
|Less developed excluding high-inflation countries||25||29||21||12||0||12||0||8||25||41||21||20|
For the period 1949–65 as a whole, 77 per cent, 50 per cent, and 60 per cent, respectively, of the industrial, other developed, and less developed groups of countries had average rates of inflation of 5 per cent or less. Even including the 6 high-inflation countries, the less developed group is not the most extreme one in this respect. When the high-inflation countries are excluded, the proportion of such less developed countries having average rates of inflation of 5 per cent or less rises from 60 per cent to 75 per cent, almost the same proportion as that of the industrial group.
In the first subperiod, smaller proportions of countries in the various groups had average rates of inflation of 5 per cent or less; thus 46 per cent of the industrial countries had such average rates in this subperiod. This proportion increased to 100 per cent in the second subperiod but decreased during the third subperiod to 92 per cent. Over the three subperiods, the proportion of the other developed countries that had average rates not greater than 5 per cent increased—50 per cent, 70 per cent, and 80 per cent. Although the proportion of all less developed countries having average rates of inflation of 5 per cent or less increased in the second subperiod compared with the first, it remained stable in the third. This was also true of the more limited group of less developed countries. Of interest in Table 2 is the fact that the proportions of the less developed countries having Relative Price Stability in the period 1949–65 as well as in the three subperiods are higher than the proportions for all countries combined. In fact, in all except the second subperiod the proportions of this group of countries having Relative Price Stability are higher than the corresponding proportions of both the industrial and the other developed groups. When the high-inflation countries are excluded, this statement becomes true for the second subperiod as well.
Table 3 presents the average rates of inflation in the 53 countries for the whole of the period 1949–65 as well as for the three subperiods. These figures form the basis for the observations made about the comparative rates of inflation in the different groups of countries. It might be instructive to look at the pattern of the individual rates that go to make up the group averages. In the industrial countries, the average rates range from 6.51 per cent in Austria to 1.63 per cent in both Switzerland and the United States. In the first two subperiods, Switzerland recorded the lowest average rates of 0.70 per cent and 1.05 per cent, respectively. The United States had the lowest rate in the third subperiod. In the three subperiods, the highest average rates were recorded, respectively, by Austria (14.28 per cent), France (4.29 per cent), and Japan (5.85 per cent). It is interesting that the North American countries (and the United Kingdom, except in the second subperiod) had average rates that were in all instances lower than those recorded, on the average, by the industrial continental Western European countries. Except in the second subperiod, Japan’s average rates were higher than those of the industrial continental Western European countries.
|Other developed countries||Australia||12.58||Turkey||14.10|
|Less developed countries||Korea||53.81||Korea||134.90||Bolivia||86.96||Brazil||58.04|
|Other developed countries||Iceland||7.75||Greece||9.69||Spain||7.15||Iceland||9.35|
|Less developed countries||Israel||9.47||Nicaragua||9.45||Colombia||8.58||Peru||9.20|
|Other developed countries||Finland||4.61||Spain||4.99||Finland||4.71||Finland||4.82|
|South Africa||2.52||New Zealand||2.28|
|Less developed countries||Tunisia||4.99||Costa Rica||4.75||Thailand||3.16||Philippines||4.67|
|Relative Price Stability|
|United States||1.63||Switzerland||0.70||Japan||1.59||United States||1.28|
|Other developed countries||Portugal||1.49||Portugal||0.65||Portugal||1.21||Greece||1.73|
|Less developed countries||Venezuela||1.66||Ceylon||1.99||Costa Rica||1.97||Tunisia||1.87|
For the other developed countries not only were the average rates themselves on the whole higher than those attained by the industrial group but the range of the average rates was wider. For the whole period, the average rates range from 1.49 per cent for Portugal to 7.75 per cent for Iceland. The average rates for this group for the full period and for the three subperiods are 4.85, 6.06, 4.79, and 3.91 per cent, respectively. The other developed countries in Europe had rates that were, on the whole, higher than the averages for all other developed countries: 5.16, 5.20, 5.64, and 4.66 per cent. This, of course, means that between them Australia, New Zealand, and South Africa had average rates that were (in all but the first subperiod) lower than the group averages—4.12, 8.06, 2.81, and 2.17 per cent.
The less developed group includes the 6 high-inflation countries— Korea, Bolivia, Chile, Brazil, Argentina, and Uruguay. Questions have been raised about the meaningfulness of any measure of average tendency that includes such extreme rates. In Table 3, it will be noted that for the whole period there is a distinct discontinuity between the rates in these 6 countries and those in the remaining 47 countries. The 19.24 per cent recorded by Uruguay (sixth in order of average rates of inflation) is more than twice that recorded by Israel, which ranks seventh.
A brief look at the underlying regional patterns of the less developed group’s average rates may be revealing. There are 17 Latin-American countries in this group. For the period as a whole, Bolivia with an average rate of 44 per cent had the highest rate and Panama with practically no change had the lowest rate of inflation. The average rates for the Latin-American subgroup for the period and subperiods studied are 11.7, 9.8, 14.3, and 10.6 per cent, respectively. If in calculating the averages for the subgroup one eliminates the observations for the 5 high-inflation countries in Latin America—Argentina, Bolivia, Brazil, Chile, and Uruguay—the corresponding average rates fall sharply, to 3.6, 5.3, 2.9, and 3.0 per cent. These rates compare favorably with those recorded by any other group or subgroup.
The behavior of prices in Korea has a similar distorting effect on the averages for the Asian subgroup of less developed countries. For the subgroup inclusive of Korea the 1949–65, 1949–53, 1954–59, and 1960–65 average rates are 10.9, 24.5, 5.6, and 5.0 per cent, respectively. Exclusive of Korea, these become much lower—2.4, 2.4, 1.6, and 3.0 per cent, respectively. An interesting point about the rates for this less inclusive subgroup is that unlike other groups and subgroups the highest average rate did not occur in the first subperiod. The relatively small number of countries in the African and Middle Eastern subgroups makes comments about average rates in these subgroups less firm. The average rates for the Middle Eastern subgroup are 3.7, 3.8, 3.8, and 3.5 per cent, respectively. For the African subgroup the average rates are 5.8, 10.3, 2.5, and5.5 percent, respectively.
In summary, then, any contention that the less developed group of countries had higher rates of inflation than the other groups depends, for the period 1949–65, solely on the extreme behavior of prices in a small proportion of the countries in this group. When these few countries are excluded, the less developed group emerges as the group with the best record. The other developed group emerges as the group with the highest average rates of inflation. Even without excluding the high-inflation countries, a higher proportion of the less developed than of any other group of countries had Relative Price Stability during 1949–65. When the high-inflation countries are excluded, a higher proportion of less developed countries had Relative Price Stability in each of the subperiods as well. Thus, the evidence raises serious questions, to say the least, about the rather widespread, though sometimes vaguely held, belief that inflation has been rampant in less developed countries.5
II. Variability of Rates of Inflation
In evaluating the average rates of inflation, it is necessary to study the variability of rates of inflation in different countries. This is done here primarily in terms of the dispersion of the various annual rates that make up the average in each country. However, the extent to which countries have actually had negative rates of inflation and the extent to which rates of inflation have tended to increase or decrease continuously are also examined briefly.
The coefficient of variation is used as the measure of the dispersion of annual rates about the average. This statistic, for each country, represents the standard deviation of the annual price increases divided by the average. The higher the coefficient, the more variable the price changes during the period covered. With the respective averages acting as scale factors, these coefficients are likely to be meaningful indicators of the comparative degrees of variability of the annual rates in the different countries. The individual country coefficients are presented in the Appendix (Table 10).
It is frequently said that the less developed economies are subject to relatively larger degrees of fluctuation or instability than others. The coefficients of variation of the rates of inflation in the three groups of countries are examined below in different ways to see whether price changes were more variable in the less developed than in the other groups of countries during the period covered.
When the countries are ranked for the period 1949–65 as well as for the three subperiods starting with the country with the highest coefficient of variation, the less developed countries tend to have relatively higher rankings. Of the countries having the 30 highest coefficients during the period 1949–65, less developed countries form 73 per cent. For the three subperiods, the corresponding proportions are 60 per cent for the first subperiod and 70 per cent each for the second and third subperiods. The same conclusion is reached about the relative degrees of the variability of price movements in the less developed countries compared with industrial and other developed countries, if one compares the group average coefficients of variation (Table 4). There, it will be seen that the average coefficients for the less developed group are almost invariably higher than those of the developed countries.
|Over 2 Per|
Cent But Not
5 Per Cent
|Over 5 Per|
Cent But Not
10 Per Cent
|Less developed||2.591||1.54 2||0.96||0.94||1.51 3|
|All countries||2.10 1||1.19 2||0.98||0.94||1.28 3|
Excludes Panama and Iraq for reasons explained in the text, page 542.
Excludes Iran for similar reasons.
Excludes Panama, Iraq, and Iran.
Excludes Panama and Iraq for reasons explained in the text, page 542.
Excludes Iran for similar reasons.
Excludes Panama, Iraq, and Iran.
The degree of confidence with which one can say that annual price movements are more variable in the less developed countries is reduced if the 6 high-inflation countries are excluded. The average coefficients of variation for the high-inflation countries (0.94, 0.68, 0.69, and 0.48) are much lower than those (4.89, 1.97, 2.28, and 1.64) obtained for the less developed countries exclusive of the high-inflation countries. The average coefficients for the less developed group inclusive of the high-inflation countries (4.10, 1.72, 1.96, and 1.40) are thus somewhat lower than those for the limited group. In fact, the high-inflation group turned out to be the group with the smallest year-to-year variability in price changes. While this may be expected, it is likely that this tendency for the high-inflation countries to have lower variability is a reflection of the way the coefficient of variation is calculated. In obtaining the coefficient, the standard deviation is divided by the average; therefore, the lower (higher) the average, the higher (lower) the coefficient that will result from a given standard deviation. The coefficient measures the deviation in relation to the average. As such, the measure makes sense. But there is something unsettling about the fact that the measure equates the degrees of instability reflected in, say, price increases varying between zero and 2 per cent, between zero and 6 per cent, and between zero and 20 per cent, if the respective average rates of price increase are 1, 3, and 10 per cent.
It may, therefore, be more meaningful to compare the degrees of average variability of prices in the different groups for average price increases in the same range. In Table 4, the average coefficients of variation for the different groups of countries are presented for each range of inflation. For this purpose, the determination of the range of inflation that a country falls into is made independently for each period or subperiod. Following the practice of eliminating observations that obviously reduce the meaningfulness of group averages, the average coefficient for the less developed group of countries with average rates of inflation of less than 2 per cent for the period 1949–65 does not include observations for 2 countries. These 2 countries—Panama and Iraq—had average rates of inflation that were very small, so that when they are used as denominators in calculating the coefficient of variation, extremely high coefficients result. For similar reasons, Iran is excluded in calculating the group average coefficient of variation for less developed countries with an average rate of inflation of between 2 per cent and 5 per cent during the period 1949–65. Leaving these coefficients out does not, however, alter the conclusions about the over-all relationships between the coefficients for the different groups of countries.
Before comparing the group coefficients, it should be noted that the coefficients do indeed display the tendency to be smaller, the higher the range of inflation. It is clear from an inspection of the table that any conclusions that may be reached about the group coefficients will not be independent of ranges of inflation. First, in the two highest ranges of inflation—i.e., over 5 per cent but not more than 10 per cent, and over 10 per cent—the differences between the group coefficients are probably not statistically significant. Although there are some subperiods and ranges in which some groups are not represented, it is probably safe to say that at the higher rates of inflation, the variability of price does not differ much among the different groups of countries.
In the two lower ranges of inflation, however, the less developed countries seem to have greater year-to-year variability in price changes. The only subperiod in which prices (in these two ranges) in the other groups of countries displayed as much variability as in the less developed group is the first subperiod. However, it should be noted that this sub-period includes the period of the Korean conflict and witnessed relatively wide fluctuations in prices, especially in the industrial and other developed countries. On the whole, it is clear that in these lower ranges of inflation, price changes in the less developed countries are subject to greater variability. The low average rates of inflation in these countries result from annual rates of inflation that vary more widely than those in the other two groups. For example, for the whole period all the 8 less developed countries with average price increases of not more than 2 per cent experienced at least four negative annual rates of change in prices. Of the 4 countries in the developed group that had average rates in the same range, only 1 had four negative annual price changes. Similar observations can be made about the subperiods. For instance, in the subperiod 1954–59, 6 of the 7 less developed countries that had average rates between 2 per cent and 5 per cent experienced at least one negative change in prices. Only 1 of 7 industrial countries and 1 of 6 other developed countries with average rates in the same range in the same subperiod experienced a similar fall in the price level. These differences in price variability at the two lower ranges between the less developed and the other groups may be a reflection of the greater susceptibility of the less developed countries to exogenous factors and of the fact that their monetary and fiscal instruments are probably less capable of bringing about marginal adjustments in the economy.
Negative rates of inflation during the period under review can be examined in more general terms. This study includes 30 less developed countries with 17 annual rates of change each, totaling 510 annual rates of change. Of these rates of change, 80 (15.7 per cent) are negative. Of the 391 rates of change for the industrial and other developed groups taken together, 24 (6.1 per cent) are negative. The proportions of the negative rates of change for the industrial and other developed countries are, respectively, 6.8 per cent and 4.1 per cent.
Looked at from another point of view, 87 per cent of the less developed countries record at least one negative rate of inflation during 1949–65, 57 per cent had at least two negative rates of inflation, 50 per cent at least three negative rates of inflation, 40 per cent at least four negative rates, 20 per cent at least five negative rates, 10 per cent at least six negative rates, and 3 per cent seven annual negative rates. Of the industrial and other developed countries together, 57 per cent record at least one negative rate of inflation, 26 per cent had at least two negative rates, and 9 per cent record four or five negative rates of inflation (one observation in each category). Unlike the less developed countries, none of the industrial and other developed countries had six or seven negative rates of inflation during the period. Also, while 43 per cent of the industrial and other developed countries together record no negative rates at all, only 13 per cent of the less developed countries had no negative rates of inflation. While 96 per cent of all the observed negative rates in the industrial and other developed countries occurred before 1956, the corresponding proportion for the less developed countries was 56 per cent. In fact, none of the industrial and other developed countries recorded a negative annual change in the cost of living index during 1960–65. In the less developed countries, 20 per cent of the negative price changes observed during 1949–65 occurred during 1960–65.
The maximum observed lengths of successive increases or decreases in the rates of inflation may now be examined. In most countries (42 or 43 out of 53), the maximum number of years of continuous increases or decreases in the rate of inflation is limited to from two to three years (Table 5). Thus, relatively few countries record accelerations or decelerations in inflation that last for more than three years.6 In 2 of these, Argentina (1955–59) and the Netherlands (1961–65), successive increases lasting for five years were observed. Successive decelerations in inflation lasting four years were observed in 4 countries,7 while in India (1957–61) and the United Kingdom (1956–60) decreases lasting five years were observed. In New Zealand, successive decreases occurred in the six-year period 1952–57.
|Mean in Years||2.6||2.6||2.7||2.7||2.5||3.3||2.6||2.7|
It will also be observed in Table 5 that the means of the maximum unidirectional upward movements in the three groups of countries are not too different. The means of the maximum successive decelerations show greater divergence. The other developed group with a mean of 3.3 years shows the longest deceleration, followed by the less developed group with 2.6 years and the industrial group with 2.5 years.
III. Trends in Inflation
One important aspect of price changes in an economy is the trend described by these changes, for a community’s expectation about future price movements is related in an important way to the past behavior of these movements. The estimated trends are linear. Trends are estimated not only for the actual annual rates but also for moving averages. An examination of the series suggests that most of them are such that linear trends may not be easily fitted to the actual rates. Like most time series, they show signs of cyclical and irregular movements in addition to any underlying trends that may be present. This was to be expected. As cyclical and other developments are mirrored in price movements, increases in the rate of inflation are likely to be followed by decreases, and so on. The comments made in Section II about the tendencies for continuous accelerations and decelerations suggest that a five-year moving average of the rates may provide series that are relatively free from cyclical and erratic movements. In spite of such smoothing, however, it is possible that in particular countries a few extreme observations will bias the estimates.
The detailed trend estimates are presented in Table 11 in the Appendix. In studying the estimates for the entire period obtained from the actual rates, it will be observed that the directions of the trend in the rates of inflation (the signs of the slope or correlation coefficients) are about equally divided between negative and positive—28 and 25, respectively.
However, in only 14 cases do tests performed at the 95 per cent level of confidence suggest that these coefficients are significantly different from zero. In Table 6, the countries are listed whose slope or correlation estimates for the entire period are statistically different from zero. In 9 of these—Australia, Austria, El Salvador, Greece, Guatemala, Korea, New Zealand, South Africa, and Tunisia—the trend in rates of inflation during 1949–65 was negative. In the remaining 5 countries—Brazil, India, the Philippines, Portugal, and Uruguay—the trend in rates of inflation during the period was positive.
|El Salvador||El Salvador|
|New Zealand||New Zealand|
|South Africa||South Africa|
As was to be expected owing to the prevalence of cyclical changes, the number of countries for which the coefficients turn out to be significantly different from zero, at the 95 per cent level of confidence, increases noticeably when the five-year moving averages of the original rates are used for trend fitting. Of the 53 countries, 27 now show trends in their rates of inflation; 19 of these 27 countries had negative trends, and 8 had positive trends. About the same proportion of less developed countries and of industrial and other developed countries had negative trends in inflation. A greater proportion of the less developed group, however, had positive trends.
The finding that most of the trends in inflation during the period 1949–65 were negative of course reflects the fact that most of the countries studied experienced their highest rates of inflation during the earlier years of the series. This is especially so for the industrial group of countries. In Table 9 in the Appendix, the annual rates of inflation for the period 1949–65 are presented for each country. Of the 53 countries, 37 recorded their highest rates of inflation in one or another of the first five years. In fact, 21 countries recorded their highest annual increases in the cost of living index in 1951. Of the 13 industrial countries, 11 had their highest rates of increase in 1951. The exceptions are Japan (1949) and the United Kingdom (1952). Although 1951 was also the most common year of highest price increases in the other developed and the less developed groups of countries, the proportions of these groups of countries in which the highest rates were observed in 1951 are lower.
One way of examining the influence of the movements in prices in the earlier years on the trend estimates for the whole period is to eliminate the early years and to estimate the trends for the remaining period. For this purpose, the years 1949–53 were eliminated from the series. This leaves 12 observations, and because of the cost in degrees of freedom, no five-year moving average estimates were made. Instead, trends were fitted to three-year moving averages. While it may thus not be possible to learn as much about the trends during this subperiod as in the period 1949–65, enough emerges to suggest that most of the negative trends found for the whole period were due to the unusually high rates of inflation that prevailed in the period 1949–53.
Preliminary support for this conclusion is obtained when the signs of the estimated coefficients, statistically significant or not, of the actual rates for the two periods are compared. For 19 countries, the signs of the trend coefficients differ. In 15 countries, while the signs of the trend coefficients for the period 1949–65 were negative, those for the sub-period 1954–65 were positive. These 15 countries, 10 of which are industrial or other developed, are Austria, Canada, Ceylon, France, Honduras, Ireland, Japan, Morocco, the Netherlands, Norway, Sweden, Tunisia, the United Kingdom, the United States, and Venezuela in the remaining 4 countries—Chile, Iran, Iraq, and Turkey—while the trend coefficients for the whole period were positive, those for the subperiod were negative.
Indeed, when attention is turned to the countries where trend coefficients estimated for the period 1954–65 meet the test of statistical significance, a picture emerges that is different from that obtained for the whole period. When trends are fitted to the three-year moving averages for the period 1954–65, 21 countries have coefficients that are statistically significant (Table 7). Seven of these countries have trends that are negative. Fourteen have positive trends. This result should be compared with the one obtained from the three-year moving averages for the period 1949–65. It is apparent from Table 7 that the basic tendencies differ. For the longer period, 16 countries have significantly negative trends and 7, significantly positive ones. Thus, while during the period 1949–65 about 70 per cent of all the statistically significant trends were negative, only 33 per cent of those estimated for the period 1954–65 were negative.
When one compares the trend tendencies of the alternative groups of countries in the 1949–65 and 1954–65 periods, one reaches conclusions that are similar to those reached when the average rates in the 1949–53, 1954–59, and 1960–65 subperiods are compared. That is, the tendency for the rates of inflation to increase in the latter part of the period covered is more noticeable in the industrial than in any other group (Table 8).
in Countries with
in the Respective Groups
Having Significant Trends
|Other developed||4||1||Other developed||40.0||10.0|
|Less developed||8||5||Less developed||26.7||16.7|
|Other developed||4||2||Other developed||40.0||20.0|
|Less developed||3||8||Less developed||10.0||26.7|
For the period 1949–65, about 27 per cent of the less developed countries had inflation trends that were negative and statistically significant, while 17 per cent had positive trends. In the period 1954–65, the corresponding proportions were 10 per cent and 27 per cent, respectively. While the proportions of the other developed countries for which statistically significant negative trends were obtained remained the same in the two periods, the proportion of countries in this group with positive trend in the period 1954–65 was twice the one obtained for the period 1949–65. The most interesting divergences between the proportions of the two periods are, however, observed in the industrial countries. While about 31 per cent of the countries in this group had negative trends in the period 1949–65, no significant negative trends were estimated for the shorter period. At the same time, the proportion of the countries in this group with significant positive trends increased fourfold in the period 1954–65 compared with the period 1949–65.
This study set out to examine certain aspects of inflation in the postwar period by comparing the behavior of price changes in a number of countries classified into three groups—industrial, other developed, and less developed. Rates of price changes for these countries are examined for the period 1949–65 and the three subperiods into which it is divided—1949–53,1954–59, and 1960–65.
The group average rates of inflation for the whole period and the three subperiods show that, when certain high-inflation countries are excluded, the less developed group of countries had, in general, lower rates of inflation than the other two groups. The group of other developed countries had the highest average rates of inflation. Even without excluding the high-inflation countries, a higher proportion of the less developed countries than of any other group had Relative Price Stability, i.e., average rates of inflation not greater than 2 per cent. In the industrial group, the North American subgroup (and the United Kingdom, except in the second subperiod) had average rates that were lower than those recorded by the continental Western European subgroup.
It is usually said that less developed economies are subject to greater instability than others. As important economic indicators, price movements may reflect this difference. Using the coefficient of variation as the measure of variability, it was found that for average rates of inflation greater than 5 per cent per annum, differences between the group coefficients were not very large. However, for average rates of inflation not greater than 5 per cent, price changes in the less developed countries showed greater variability than in the other groups of countries. One other aspect of variability is indicated by the relative occurrence of negative price changes. More negative rates of change of prices were observed in the less developed than in the other groups of countries.
Negative price changes constituted 16 per cent of all the annual price changes in the less developed countries, while they constituted only 6 per cent of those in the industrial and other developed countries. While about 30 per cent of the negative price changes in the less developed countries occurred during 1960–65, no industrial or other developed country had negative price changes during that period.
The final aspect of inflation that was examined was the behavior of the trend. During the period 1949–65, trends fitted to five-year moving averages of the rates of inflation were statistically significant in 27 of the 53 countries. In 19 of these 27 countries, the trends were negative. About the same proportion of less developed and industrial and other developed countries had negative rates of inflation. However, a greater proportion of the less developed group had positive trends in inflation. The predominance of negative trends during the period 1949–65, however, seems to reflect the fact that most countries, under the influence of the Korean conflict, experienced their highest rates of inflation in the earlier years of the series. When trends are fitted to the observations for the period 1954–65, the over-all tendency in the 21 countries in which statistically significant trends were found was for positive trends. This tendency was more noticeable in the industrial than in any of the other groups of countries.
|United Arab Republic||—1.08||5.43||9.28||—0.94||—6.67||—4.08||0||2.13||4.17||0||0||1.00||0||—2.97||1.02||4.04||14.56|
|Other developed countries|
|Less developed countries|
|Iran||(—) 11.72||(—) 1.91||2.34||1.07|
|United Arab Republic||3.17||4.62||6.80||1.90|
|Actual Rates||3-Year Moving|
|Actual Rates||3-Year Moving|
|b (t value)|
|b (t value)|
|United Arab Republic||0.234||0.236||0.038||0.059||0.129||0.385||0.696||0.532||0.212||0.297|
Coefficient is significant at the 5 per cent level. The significance tests for the b and r coefficients are related.
p1 = a + bt + u .
Coefficient is significant at the 5 per cent level. The significance tests for the b and r coefficients are related.
p1 = a + bt + u .
Le rythrne de l’inflation dans les pays industriéis, les autres pays développés et les pays moins développés, 1949–65
Cette étude analyse les taux moyens d’inflation, leur variabilité annuelle et la tendance qui s’en dégage dans 53 pays réparts en trois groupes — pays industriels, autres pays développés et pays moins développés — au cours de la période 1949–65 ainsi que pendant trois sous-périodes. Pour mesurer l’inflation, on s’est servi de l’indice du coût de la vie.
Le groupe des pays moins développés a enregistré un taux moyen d’inflation moins élevé que les autres groupes si l’on exclut six pays ayant connu une inflation rapide : l’Argentine, la Bolivie, le Brésil, le Chili, la Coréé et l’Uruguay. Méme si on les inclut, la stabilité des prix se rencontre plus fréquemment dans le groupe des pays moins développés que dans tout autre groupe.
La mesure de la variabilité du taux d’inflation dans un pays donné au cours de la période considérée a été donnée par le coefficient de variation. Si l’on considère les pays ayant un taux moyen d’inflation élevé, on constate que la variabilité de l’inflation est la méme dans les pays moins développés que dans les autres. Toutefois, en ce qui concerne les pays ayant un rythme d’inflation lent, on enregistre une variabilité plus élevée dans les pays moins développés que dans les autres. Cette différence s’explique en partie par le fait, que, dans le groupe des pays ayant un taux moyen d’inflation faible, les pays moins développés ont connu des années de fléchissement des prix plus fréquentes que les autres pays.
Du point de vue statistique, on n’a pu dégager de tendance significative des taux d’inflation que dans la moitié des pays; en ce qui concerne ceux-ci, la majorité a enregistré un ralentissement du rythme de l’inflation. Ce ralentissement est aussi fréquent chez les pays moins développés que dans les autres pays, mais les cas d’accélération de l’inflation sont plus fréquents chez les premiers. Si l’on ne tient pas compte de la période couverte par le conflit coréen, caractérisée par une montune montée des prix, la proportion des cas de taux croissants d’inflation au cours de la période augmente. Cette fréquence accrue des cas d’inflation croissante s’observe surtout chez les pays industrialisés, mais également chez les pays moins développés.
La tasa de inflación en los países industriales, otros desarrollados, y menos desarrollados, 1949–65
En este estudio se examinan la tasa media de inflación, la variación de esa tasa de un año a otro, y las tendencias en las mismas durante el período 1949–65 y también para tres subperíodos. Los datos se refieren a 53 países clasificados en tres grupos: industriales, otros desarrollados, y menos desarrollados. Se utilizó el índice del costo de vida como medida de la inflación.
Al excluir del grupo menos desarrollado a los seis países de alta inflación—Argentina, Bolivia, Brasil, Corea, Chile y Uruguay—se obtuvo una tasa media de inflación inferior a la de los otros grupos. Aun cuando se incluyó a esos países en el grupo, hubo una mayor proporción de estabilidad de precios entre los países menos desarrollados que en ningún otro grupo.
La variabilidad de las tasas de inflación en un país dado a lo largo del tiempo se midió mediante el coeficiente de variación. Entre los países con una alta tasa media de inflación, los países menos desarrollados mostraron aproximadamente la misma variabilidad de inflación que los otros. Pero entre los países con bajas tasas de inflación, los países menos desarrollados mostraron mayor variabilidad que los otros. Relacionado con esta diferencia se encuentra el hecho de que, entre los países que tenían bajas tasas medias de inflación, los países menos desarrollados experimentaron más años de descenso de precios que los otros países.
Solamente se pudo encontrar tendencias estadísticamente significativas en las tasas de inflación en la mitad de los países, y dentro de esa mitad la mayoría experimentó disminuciones en su tasa de inflación. Los países menos desarrollados registraron la misma frecuencia de tasas decrecientes de inflación que los otros países, pero una mayor frecuencia de tasas crecientes de inflación. Cuando se excluye el período de precios elevados que vino asociado al conflicto coreano, se eleva la proporción de casos de tasas crecientes de inflación a lo largo del tiempo. Especialmente entre los países industrializados, pero también entre los países menos desarrollados, se encuentra una aumentada frecuencia de inflación creciente.
Mr. Adekunle, a graduate of Ohio Wesleyan University and of the University of Wisconsin, was an economist in the Fund’s Research Department when this paper was prepared. He is now employed by the Central Bank of Nigeria.
The cost of living index numbers used here are those contained in the International Monetary Fund’s publication, International Financial Statistics.
Graeme S. Dorrance, “Inflation and Growth: The Statistical Evidence,” Staff Papers, Vol. XIII (1966), p. 98.
Ibid. See also U Tun Wai, “The Relation Between Inflation and Economic Development: A Statistical Inductive Study,” Staff Papers, Vol. VII (1959-60), pp. 311-12.
Paraguay, which has also had an extremely high rate of inflation, is not covered in this study because data for it are not available for the entire period.
See, e.g., Enke’s unequivocal statement that “there are few under-developed countries that do not experience at least a 5 per cent inflation of price levels each year. And annual inflation rates of 20 per cent are not uncommon,” Stephen Enke, Economics for Development (Englewood Cliffs, N.J., 1963), p. 219.
Argentina, Brazil, Iceland, Japan, Morocco, Netherlands, Spain, and United Arab Republic.
Bolivia, Finland, Greece, and Guatemala.