1. How important is the permanent component of price shocks?

As noted earlier, the design and feasibility of stabilization and hedging strategies depend importantly on the nature of the shock. At the risk of oversimplifying, it can be said that income stabilization policies and hedging are useful in dealing only with temporary and, preferably, short-lived shocks. Permanent shocks require adjustment and, possibly, the implementation of structural policies. To determine how important a role, if any, permanent shocks play in explaining the variability in commodity prices, we proceed in two steps. First, we establish the time series properties of the various commodity indices via the standard Augmented Dickey-Fuller (ADF) and the Phillips-Perron unit root tests under two alternative hypotheses: the first of these assumes no structural breaks occurred within the sample; the second allows for a one-time break in both the mean and the trend at a pre-specified point in time (see Perron (1988)). ^5/ If the series is stationary, all shocks are temporary and the null hypothesis of a unit root is rejected. If the null hypothesis is not rejected, then the series has a permanent component of unknown size. Second, to determine the size of the permanent component we employ Cochrane’s (1988) methodology, which provides a measure of the persistence of shocks to a variable by examining the variance of its long differences.

The form of the ADF test employed, given by equation (1), allows for both the presence of a constant (nonzero mean) and a constant deterministic drift. To ensure that the regression residuals are serially uncorrelated, past differences of the variable (denoted by Δ) are also included. As suggested by Campbell and Perron (1991), we begin by including a generous number of lags; if the past differences do not enter significantly, these are dropped sequentially. ^6/ Similarly, the Phillips-Perron test allows for constant and drift terms.

The results of the unit root tests, which are summarized in Table 3, uniformly indicate that, when no breaks are assumed, for the log of the price series examined the null hypothesis of a unit root could not be rejected at standard confidence levels. ^7/ Perron (1989), however, has suggested that it is important to model large “shocks” which may result in a one-time break in either the level, the slope (or both of these) of macroeconomic time series when testing for unit roots. Standard unit root tests, Perron (1988) shows, have little power to distinguish between nonstationarity and stationarity around a broken trend. The 1973 oil price shock, for instance, may have had persistent effects on real commodity prices.

Table 3.

The Unit Root Tests 1957:I-1993:II

^1/The critical values are taken from Guilkey and Schmidt (1989). The number of observations in the sample period is 146.

^2/These are taken from Perron (1989). Chart A.1 plots the fitted broken trends to the data.

Table 3.

The Unit Root Tests 1957:I-1993:II

Alternative hypothesis: no structural breaks
Regression: $\begin{array}{c}\mathrm{\Delta }{\mathit{y}}_t=\mu +\mathit{\beta t}+\mathit{\alpha }{\mathit{y}}_{t-1}+\underset{i=1}{\overset{k}{\mathrm{\Sigma }}}{\mathit{\delta \Delta y}}_{t-i}+e_t\end{array}$
Series	k	t-statistic on α (A.D.F. statistic)	Phillips-Perron statistic
All commodities	5	-2.1353	-2.6112
Beverages	5	-2.5308	-1.4472
Food	3	-1.8665	-2.4743
Metals	1	-3.3555	-3.2849
Critical values	1%	5%	10%
150 observations 1/	-3.73	-3.46	-3.16

Alternative hypothesis: one-time structural break occurring at TB
Alternative Hypothesis: yt = μ1 + β1t + (μ2 - μ1)DUt (β2-β1)DTt + et
Regression: $\begin{array}{c}\mathrm{\Delta }{\hat{\mathit{y}}}_t=\mathit{\alpha }{\hat{\mathit{y}}}_{t-1}+\underset{i=1}{\overset{k}{\mathrm{\Sigma }}}\mathit{\delta \Delta }{\hat{\mathit{y}}}_{t-i}+e_t\end{array}$
Series	TB	t-statistic α (A.D.F.)	Phillips-Perron	λ	Critical values 2/
					1%	5%	10%
All commodities	73:I	-6.449	-4.306	0.4	-4.81	-4.22	-3.95
Beverages	76:I	-3.353	-3.955	0.5	-4.90	-4.24	-3.96
Food	73:I	-4.082	-3.949	0.4	-4.81	-4.22	-3.95
Metals	74:III	-4.558	-3.659	0.5	-4.90	-4.24	-3.96

^1/The critical values are taken from Guilkey and Schmidt (1989). The number of observations in the sample period is 146.

^2/These are taken from Perron (1989). Chart A.1 plots the fitted broken trends to the data.

View Table

Table 3.

The Unit Root Tests 1957:I-1993:II

Alternative hypothesis: no structural breaks
Regression: $\begin{array}{c}\mathrm{\Delta }{\mathit{y}}_t=\mu +\mathit{\beta t}+\mathit{\alpha }{\mathit{y}}_{t-1}+\underset{i=1}{\overset{k}{\mathrm{\Sigma }}}{\mathit{\delta \Delta y}}_{t-i}+e_t\end{array}$
Series	k	t-statistic on α (A.D.F. statistic)	Phillips-Perron statistic
All commodities	5	-2.1353	-2.6112
Beverages	5	-2.5308	-1.4472
Food	3	-1.8665	-2.4743
Metals	1	-3.3555	-3.2849
Critical values	1%	5%	10%
150 observations 1/	-3.73	-3.46	-3.16

Alternative hypothesis: one-time structural break occurring at TB
Alternative Hypothesis: yt = μ1 + β1t + (μ2 - μ1)DUt (β2-β1)DTt + et
Regression: $\begin{array}{c}\mathrm{\Delta }{\hat{\mathit{y}}}_t=\mathit{\alpha }{\hat{\mathit{y}}}_{t-1}+\underset{i=1}{\overset{k}{\mathrm{\Sigma }}}\mathit{\delta \Delta }{\hat{\mathit{y}}}_{t-i}+e_t\end{array}$
Series	TB	t-statistic α (A.D.F.)	Phillips-Perron	λ	Critical values 2/
					1%	5%	10%
All commodities	73:I	-6.449	-4.306	0.4	-4.81	-4.22	-3.95
Beverages	76:I	-3.353	-3.955	0.5	-4.90	-4.24	-3.96
Food	73:I	-4.082	-3.949	0.4	-4.81	-4.22	-3.95
Metals	74:III	-4.558	-3.659	0.5	-4.90	-4.24	-3.96

^1/The critical values are taken from Guilkey and Schmidt (1989). The number of observations in the sample period is 146.

^2/These are taken from Perron (1989). Chart A.1 plots the fitted broken trends to the data.

To examine this possibility, the deterministic component was modelled to incorporate a single structural break, at which time both the level and slope of the deterministic component changed (the exact specification used is shown in the bottom panel of Table 3). As in Perron (1988) the sample behavior of the data was allowed to indicate where the “potential” break, denoted in the Table by T_B, occurred. These breaks are timed closely with the oil shock for the all-commodities, food, and metals baskets. For beverages the 1977 coffee supply shock appears as the best candidate of a behavioral change. The actual prices and the deterministic broken trends are plotted in Chart A.1. As Table 3 shows, the results of the Perron unit root tests indicate that the metal index is stationary around the broken trend while for the food and beverages indices the unit root hypothesis cannot be rejected. For the all-commodity index the results are ambiguous, but the more powerful Phillips-Perron tests indicate nonstationarity. However, the power of the Perron tests has recently come into question (see Zivot and Andrews (1992)) and their analysis shows these results should be interpreted with care. Given that: (a) there are ambiguities about the exact time series properties of some of the commodity price indices (all-commodities and metals), (b) the more general concerns about the power of the Perron test, (c) the lack of consensus in the literature about the existence and timing of breaks, we proceed under the hypothesis that the series in question have a unit root component. ^8/9/

However, as the unit root tests are silent as to the size and relative importance of the permanent component, we proceed to shed light on this issue by employing the methodology proposed by Cochrane (1988). Specifically, if the variable y has the following representation,

then if δ=1 and the disturbance term, ε, is white noise, then y follows a random walk and the variance of its k-differences grows linearly with the difference

If δ<1, so y is a stationary process, the variance of its k-differences is given by

Therefore, the ratio (1/k)var(y_t-y_t-k)/var(y_t-y_t-1) is equal to one if y is a random walk. If y is stationary all shocks eventually die out, so the variance ratio converges to zero. If y is a more general I(1) process, which has both permanent and temporary (stationary) components, the ratio will converge to the ratio of the variance of the permanent shock to the total variance of y. Thus, the closer that ratio is to unity, the larger the size of the unit root component and the lower is the relative importance of temporary shocks. In the limiting case, where the ratio converges to unity, there would be no scope for stabilization policies since all shocks are permanent. The presence of an important temporary component, however, does not guarantee that price (and/or income) stabilization policies will be feasible, since persistence in the temporary component may be so high so as to make stabilization extremely costly. Hence, as regard stabilization policies, “temporariness” is a necessary but not sufficient condition. The nature of and persistence in the temporary component will be examined in the following section.

Table 4 summarizes the main results. The values of k range between one and twenty years. For each of the four commodity indices, permanent shocks play a role in explaining the variance of commodity prices. However, its relative importance varies considerably across commodity groupings and just as much within groupings (see Cuddington (1992)). Permanent shocks are least important for metals (29 percent of the variance of yearly changes) and most important (85 percent) for beverages. The all commodity index, reflecting the differing characteristics of its components, falls in the middle of the range, with permanent shocks accounting for about 45 percent of the variance of yearly changes. ^10/ For the two price indices with the largest unit root component, beverages and food, the ratio took the longest number of years to converge, possibly indicating that even the temporary component of these series exhibits a high degree of persistence. The policy implications of these findings are discussed in the final section.

Table 4.

1/k Times the Variance of k-Differences ^1/

1957-92

^1/Annual data was used. All commodity price indices are deflated by the export unit values of manufactures. Standard errors are in parentheses. The standard errors were tabulated from Monte Carlo simulations (600 replications) for n=35 and k=10, 12, and 20 under the null hypothesis of a random walk.

Table 4.

1/k Times the Variance of k-Differences ^1/

1957-92

σk2/σ12 for various k (Years)
2	4	8	10	12	20
All commodities
1.032	0.465	0.524	0.448
(0.007)	(0.013)	(0.022)	(0.026)
Beverages
1.176	1.220	0.904	0.998	1.260	0.846
(0.007)	(0.013)	(0.025)	(0.029)	(0.033)	(0.039)
Food
1.188	0.816	0.682	0.613	0.731
(0.007)	(0.013)	(0.021)	(0.024)	(0.027)
Metals
1.099	0.481	0.333	0.287
(0.007)	(0.013)	(0.022)	(0.026)

^1/Annual data was used. All commodity price indices are deflated by the export unit values of manufactures. Standard errors are in parentheses. The standard errors were tabulated from Monte Carlo simulations (600 replications) for n=35 and k=10, 12, and 20 under the null hypothesis of a random walk.

View Table

Table 4.

1/k Times the Variance of k-Differences ^1/

1957-92

σk2/σ12 for various k (Years)
2	4	8	10	12	20
All commodities
1.032	0.465	0.524	0.448
(0.007)	(0.013)	(0.022)	(0.026)
Beverages
1.176	1.220	0.904	0.998	1.260	0.846
(0.007)	(0.013)	(0.025)	(0.029)	(0.033)	(0.039)
Food
1.188	0.816	0.682	0.613	0.731
(0.007)	(0.013)	(0.021)	(0.024)	(0.027)
Metals
1.099	0.481	0.333	0.287
(0.007)	(0.013)	(0.022)	(0.026)

^1/Annual data was used. All commodity price indices are deflated by the export unit values of manufactures. Standard errors are in parentheses. The standard errors were tabulated from Monte Carlo simulations (600 replications) for n=35 and k=10, 12, and 20 under the null hypothesis of a random walk.

It is worth noting that, as this methodology is based entirely on the univariate behavior of the variable, it does not discriminate among types or sources of temporary shocks. For example, it does not distinguish between a temporary shock due to a recession in the industrial countries and a temporary shock arising from a bumper crop. Such a distinction would require a full structural model of real commodity prices (see, for example, Borensztein and Reinhart (1994)).

2. The rising volatility of commodity prices

In addition to assessing whether a commodity price is driven primarily by permanent or transitory shocks, it is important to consider other aspects of the nature of the shocks when designing policy. For instance, even if the price is stationary around a deterministic trend (so all shocks are temporary), there are relatively few gains from setting up a stabilization fund and/or hedging if the variance of prices is small and large shocks are rare. The benefits that can be obtained from stabilization funds, or more generally, from precautionary savings will increase in a more volatile and uncertain environment, as greater uncertainty in a country’s export revenue stream increases the value to that country of accumulating assets as a means of insuring itself against future shocks (Ghosh and Ostry (1993)). Similarly, hedging strategies acquire greater importance as volatility increases and the probability of large destabilizing shocks is high. To examine these issues, we focus on the sample moments of the price series of interest.

To examine the issue of rising volatility, we first test for the presence of heteroskedastic disturbances and then proceed to review the descriptive statistics of commodity prices over different sample periods. The tests for heteroskedasticity on the log-differences of real commodity prices (the results for the all-commodity index are reported in Table 5) uniformly reject the null hypothesis that the shocks to commodity prices are identically distributed during the sample period considered. ^11/

Table 5.

The Changing Variance of Commodity Prices: Tests for Heteroskedasticity 1959:I-1993:II ^1/

^1/Due to the nonstationarity of the data, the tests were performed on the log-differences of the real all-commodity price index.

Table 5.

The Changing Variance of Commodity Prices: Tests for Heteroskedasticity 1959:I-1993:II ^1/

ARCH test	χ2(1) =	7.683
		(0.005)
Breusch-Pagan test	χ2(3) =	6.377
		(0.095)
White’s test	χ2(5) =	11.034
		(0.051)

^1/Due to the nonstationarity of the data, the tests were performed on the log-differences of the real all-commodity price index.

View Table

Table 5.

The Changing Variance of Commodity Prices: Tests for Heteroskedasticity 1959:I-1993:II ^1/

ARCH test	χ2(1) =	7.683
		(0.005)
Breusch-Pagan test	χ2(3) =	6.377
		(0.095)
White’s test	χ2(5) =	11.034
		(0.051)

^1/Due to the nonstationarity of the data, the tests were performed on the log-differences of the real all-commodity price index.

These results are consistent with the picture that emerges from Table 6, which presents the evolution of the basic descriptive statistics of the real commodity indices over three subsamples. Several features are worth noting. First, the average price (as well as the mode, which is not reported) is markedly lower during the most recent sample, consistent with the presence of a negative trend. ^12/ Second, there is a sustained and sharp increase in the variance of commodity prices. ^13/ This is evident in all the indices but most pronounced in the all commodities and food groupings. ^14/ The coefficient of variation rises sharply as prices become more volatile around a falling mean; for food the increase in the coefficient of variation is sixfold. Chart 2, which plots the coefficients of variation (based on a moving 15-year sample) for the 1972:I-1993:II period, also highlights the marked rise in volatility. Not surprisingly, the sharpest increases in volatility appear to have taken place during the early 1970’s following the breakdown of the Bretton Woods exchange system and on the heels of the first oil shock. However, volatility has continued to increase during the 1980s and 1990s. Lastly, the evidence on the probability of large shocks is mixed; the distributions of commodity prices (true for all indices considered) were characterized by excess kurtosis (fat tails) during the 1970s, as large shocks became more commonplace. ^15/ However, the more recent sample is characterized by relatively infrequent realizations of large shocks.

Table 6.

Descriptive Statistics 1957:I - 1993:II ^1/

^1/The variables are the logs of the commodity index deflated by the price of manufactures. The statistics are sample moments; note that due to potential nonstationarity, the population moments may not be not well defined. The numbers in parentheses are probability values under the null hypothesis of normality.

Table 6.

Descriptive Statistics 1957:I - 1993:II ^1/

Sample period	1957:I - 1969:IV	1970:I - 1979:IV	1980:I - 1993:II
Number of observations	52	40	54
All commodities
Mean	4.714	4.705	4.386
Variance	0.001	0.017	0.031
Coefficient of variation	0.789	2.806	4.036
Skewness	1.275	1.291	-0.061
	(0.000)	(0.001)	(0.859)
Kurtosis	1.939	1.019	-1.386
	(0.008)	(0.229)	(0.052)
Beverages
Mean	4.573	4.668	4.203
Variance	0.018	0.112	0.199
Coefficient of variation	2.974	7.178	10.611
Skewness	1.152	0.979	-0.411
	(0.001)	(0.015)	(0.230)
Kurtosis	0.083	0.396	-1.254
	(0.909)	(0.640)	(0.078)
Food
Mean	4.766	4.789	4.341
Variance	0.002	0.042	0.053
Coefficient of variation	0.839	4.280	5.300
Skewness	0.428	1.129	0.138
	(0.221)	(0.005)	(0.687)
Kurtosis	-0.096	0.338	-1.526
	(0.901)	(0.690)	(0.032)
Metals
Mean	4.819	4.677	4.412
Variance	0.008	0.016	0.024
Coefficient of variation	1.868	2.737	3.490
Skewness	0.619	1.299	-0.250
	(0.039)	(0.001)	(0.467)
Kurtosis	-0.280	1.638	-0.802
	(0.700)	(0.053)	(0.260)

^1/The variables are the logs of the commodity index deflated by the price of manufactures. The statistics are sample moments; note that due to potential nonstationarity, the population moments may not be not well defined. The numbers in parentheses are probability values under the null hypothesis of normality.

View Table

Table 6.

Descriptive Statistics 1957:I - 1993:II ^1/

Sample period	1957:I - 1969:IV	1970:I - 1979:IV	1980:I - 1993:II
Number of observations	52	40	54
All commodities
Mean	4.714	4.705	4.386
Variance	0.001	0.017	0.031
Coefficient of variation	0.789	2.806	4.036
Skewness	1.275	1.291	-0.061
	(0.000)	(0.001)	(0.859)
Kurtosis	1.939	1.019	-1.386
	(0.008)	(0.229)	(0.052)
Beverages
Mean	4.573	4.668	4.203
Variance	0.018	0.112	0.199
Coefficient of variation	2.974	7.178	10.611
Skewness	1.152	0.979	-0.411
	(0.001)	(0.015)	(0.230)
Kurtosis	0.083	0.396	-1.254
	(0.909)	(0.640)	(0.078)
Food
Mean	4.766	4.789	4.341
Variance	0.002	0.042	0.053
Coefficient of variation	0.839	4.280	5.300
Skewness	0.428	1.129	0.138
	(0.221)	(0.005)	(0.687)
Kurtosis	-0.096	0.338	-1.526
	(0.901)	(0.690)	(0.032)
Metals
Mean	4.819	4.677	4.412
Variance	0.008	0.016	0.024
Coefficient of variation	1.868	2.737	3.490
Skewness	0.619	1.299	-0.250
	(0.039)	(0.001)	(0.467)
Kurtosis	-0.280	1.638	-0.802
	(0.700)	(0.053)	(0.260)

^1/The variables are the logs of the commodity index deflated by the price of manufactures. The statistics are sample moments; note that due to potential nonstationarity, the population moments may not be not well defined. The numbers in parentheses are probability values under the null hypothesis of normality.

Chart 2.

The Rising Volatility in Real Commodity Prices

Coeffsicients of Variation

1972:I - 1993:II

View Expanded

Sources: International Financial Statistics, IMF and the authors.Note: The coeffcient of variation is based on fifteen-year moving averages and fifteen year standard deviations which are backward looking, hence the observation for 1972:I, for instance, is based on data from 1957:II - 1972:I.

1. Trend-cycle decompositions

If a unit root is present, then admissible ways of disentangling trend from cycle is either by the methodology proposed by Beveridge and Nelson (1981) or by the structural time series approach associated with Harvey (1985), as both of these approaches allow for the presence of a stochastic trend component. We apply, in turn, both of these approaches.

First, employing the Beveridge-Nelson (B-N) technique (Beveridge and Nelson (1981)) as modified by Miller (1988), we decompose the real commodity price into its “permanent” (or steady-state) component and “temporary” (or cyclical component), denoted by ζ and c, respectively. As discussed above, the identifying criteria for this technique is that the former captures the nonstationary component of the variable, while the latter captures its stationary element. Hence,

The evolution of the permanent component is given by

where θ and ϕ are the parameters describing the ARMA process. Using the estimates of θ, ϕ, and ε the path of the permanent component is constructed. The cyclical component is calculated residually, as the difference between the estimated permanent component, z, and the actual values of y. ^17/

Using the Box-Ljung Q statistic as a guideline, the ARMA processes were selected so as to whiten the error. In general the longer ARMA processes provided the best fit. ^18/ At the quarterly frequency, an ARMA(20,4) process was fitted to the all-commodity and beverages groupings, while food and metals were characterized by ARMA(16,4) processes. Table A.1 in the Appendix reports the full estimation results for the annual frequency and the main diagnostics of the quarterly estimation. Chart 3 plots the actual series along with the estimated permanent component. For example, large deviations from trend were evident in the all-commodity, food, and metals indices during 1973, at the time of the first oil shock. For beverages, the largest deviations from trend during the sample take place during 1977 and is associated with a supply shock. ^19/ As Chart 3 illustrates, with the exception of the metals basket, for which the actual price is below trend since late 1990, actual prices for all other indices are close to their permanent trend component. Chart 3 also illustrates that the evolution of the permanent component changes considerably during the course of the sample. While during the 1960’s and up to the first oil shock all prices were relatively stable, the trend becomes markedly negative since the mid-1970’s (for beverages somewhat later).

Chart 3.

Real Commodity Prices: Trends & Cycles Beveridge-Nelson Decomposition

1962:I - 1993:II

View Expanded

Sources: International Financial Statistics, IMF and the authors.

An alternative approach to decomposing a time series into its stochastic trend and cycle is associated with Harvey (1985). This structural time series approach relies on the Kalman filter. The stochastic trend or permanent component, z_t, can be modeled as:

where η_t and ε_t are uncorrelated white-noise disturbances.

The cycle is given by

where, as before, c_t is the cyclical component, ω_t and ω_t* are uncorrelated white noise disturbances and c_t* appears by construction (see Harrison and Akram (1983)). The parameters λ and ρ describe the frequency of the cycle and the damping factor on the amplitude of the cycle, respectively. For further details, see Harvey (1989).

The Maximum Likelihood (ML) estimates of the structural time series model present a similar scenario to the Beveridge-Nelson decomposition. Chart 4 plots the permanent component produced by this technique alongside the actual price series; as before, the bulk of the recent price weakness is associated with the secular component and there is no evidence of an abnormally large cycle. The most substantive differences between the results from the two methods are that: (a) the Kalman-filter approach produces smoother trend components for the beverages and food groupings, and, hence, a somewhat more volatile cycle and, (b) metals prices during 1993 are closer to trend using the structural time series approach than using the B-N decomposition. However, even giving the results of the Perron tests a higher weight and treating the all-commodity and metals indices as stationary about a broken trend, the main implications about trends and cycles remain. As Chart A1 in the Appendix highlights, the recent weakness does not appear to be of a primarily cyclical nature.

Chart 4.

Real Commodity Prices: Trends & Cycles Structural Time Series Decomposition

1957:I - 1993:II

View Expanded

Sources: International Financial Statistics, IMF and the authors.

2. Characteristics of the cycle

Chart 5 plots the cycles generated by the two methods; as noted earlier, the similarities are strong. In effect, the pairwise correlations between the cycles generated by the two approaches are quite high: 0.95 for the all-commodity index, 0.87 for beverages, 0.90 for food, and 0.87 for metals. ^20/ It is also evident from these charts the cycles are correlated across commodity groupings. Cycles in the all-commodity index are highly correlated (correlations above 0.80) with cycles in food and metals. Beverages exhibit the most idiosyncratic behavior, as cycles appear to be primarily driven by supply shocks.

Chart 5.

Real Commodity Price Cycles

Beveridge-Nelson Cycle (left panel) versus Kalman filter Cycle (right panel)

1957:I - 1993:II

View Expanded

Sources: International Financial Statistics, IMF and the authors.

As formerly noted, the temporary shocks that drive the cycle can exhibit differing degrees of persistence. Stabilization funds are better suited to smoothing out short-lived shocks; the more persistent the shock—the greater the costs of maintaining an “artificially” high internal price (assuming the shock had tended to depress prices). To examine this issue, we first fit an AR process to the cycle component. The Schwarz criterion was employed to select the appropriate specification (see Lütkepohl (1985) for a discussion of the advantages of this criteria over existing alternatives). Secondly, the estimated model was used to generate impulse responses that allow us to visualize the time it takes for the effects of the shock to disappear (or become arbitrarily small). Table 7 presents the estimated model while Chart 6 plots the impulse responses. While the full effects of the shock may take longer to disappear, for practical purposes we will focus on the time it takes for these effects to become smaller than some arbitrary cutoff, say plus/minus 0.02 of a unitary shock, after which the impact of the shock are minimal. Based on these considerations, it appears shocks to the metals index are the least persistent (reaching the cutoff in six to seven quarters) and most persistent for beverages (12 to 13 quarters). ^21/

Table 7.

Persistence and Dynamics of the Commodity Price Cycle ^1/

^1/Standard errors are in parentheses. There are no problems associated with nonstationarity, as the cyclical component is, by construction, a stationary, zero-mean variable. The optimum lag length for the AR processes was selected according to the Schwarz Information Criteria. Impulse responses are plotted in Chart 6.

Table 7.

Persistence and Dynamics of the Commodity Price Cycle ^1/

The dependent variable is the cyclical component of:	All Commodities	Beverages	Food	Metals
Beveridge-Nelson cycle: 1966:I-1993:II
AR(p)	AR(4)	AR(5)	AR(4)	AR(4)
AR(1)	1.296	1.278	1.279	1.206
	(0.078)	(0.091)	(0.088)	(0.094)
AR(2)	-0.795	-0.457	-0.696	-0.462
	(0.128)	(0.135)	(0.142)	(0.147)
AR(3)	0.818	0.525	0.621	0.279
	(0.131)	(0.132)	(0.143)	(0.147)
AR(4)	-0.623	-0.786	-0.429	-0.260
	(0.080)	(0.136)	(0.900)	(0.093)
AR(5)	*****	0.359	*****	*****
		(0.922)
R2	0.883	0.879	0.829	0.817
Structural Time Series cycle: 1960:I-1993:II
AR(p)	AR(3)	AR(1)	AR(3)	AR(2)
AR(1)	0.609	0.727	0.492	0.639
	(0.191)	(0.181)	(0.194)	(0.184)
AR(2)	-0.586	*****	0.322	-0.337
	(0.193)		(0.219)	(0.179)
AR(3)	0.353	*****	-0.444	*****
	(0.190)		(0.194)
R2	0.321	0.389	0.354	0.617

^1/Standard errors are in parentheses. There are no problems associated with nonstationarity, as the cyclical component is, by construction, a stationary, zero-mean variable. The optimum lag length for the AR processes was selected according to the Schwarz Information Criteria. Impulse responses are plotted in Chart 6.

View Table

Table 7.

Persistence and Dynamics of the Commodity Price Cycle ^1/

The dependent variable is the cyclical component of:	All Commodities	Beverages	Food	Metals
Beveridge-Nelson cycle: 1966:I-1993:II
AR(p)	AR(4)	AR(5)	AR(4)	AR(4)
AR(1)	1.296	1.278	1.279	1.206
	(0.078)	(0.091)	(0.088)	(0.094)
AR(2)	-0.795	-0.457	-0.696	-0.462
	(0.128)	(0.135)	(0.142)	(0.147)
AR(3)	0.818	0.525	0.621	0.279
	(0.131)	(0.132)	(0.143)	(0.147)
AR(4)	-0.623	-0.786	-0.429	-0.260
	(0.080)	(0.136)	(0.900)	(0.093)
AR(5)	*****	0.359	*****	*****
		(0.922)
R2	0.883	0.879	0.829	0.817
Structural Time Series cycle: 1960:I-1993:II
AR(p)	AR(3)	AR(1)	AR(3)	AR(2)
AR(1)	0.609	0.727	0.492	0.639
	(0.191)	(0.181)	(0.194)	(0.184)
AR(2)	-0.586	*****	0.322	-0.337
	(0.193)		(0.219)	(0.179)
AR(3)	0.353	*****	-0.444	*****
	(0.190)		(0.194)
R2	0.321	0.389	0.354	0.617

^1/Standard errors are in parentheses. There are no problems associated with nonstationarity, as the cyclical component is, by construction, a stationary, zero-mean variable. The optimum lag length for the AR processes was selected according to the Schwarz Information Criteria. Impulse responses are plotted in Chart 6.

Chart 6.

Impulse Responses: Persistence in the Cycle

Beveridge-Nelson Cycle (left panel) versus Kalman filter Cycle (right panel)

View Expanded

Notes: These impulse responses are based on the estimated AR processes reported on Table 4.

3. General observations on the time-series properties of commodity prices

A series of recent papers, including Cuddington and Urzua (1989), Powell (1991), Ardeni and Wright (1992), Deaton and Laroque (1992), and Schmitt-Grohé (1993), analyze the behavior of real commodity prices. These papers, which give careful consideration to modelling the time series properties of commodity prices (usually for long-run annual data), often present contradictory results. ^22/ There is relatively little consensus on whether real commodity prices are stationary or not or when, if any, structural breaks have taken place.

Similar ambiguities arise from the analysis of the quarterly post-war data used in this study. There is, however, one robust finding in the empirical work—real commodity prices exhibit a high degree of persistence and the high autocorrelation in the data is well documented. So while ambiguities about the exact time series properties of real commodity prices remain, a number of practical implications still emerge from an analysis of their time series behavior. First, whether the trend component is modelled by a simple broken deterministic linear trend (Char A.1 in Appendix) or by a more complicated stochastic process using alternative methodologies (this section), the recent decline in commodity prices appears to be primarily of a secular nature (Charts 3, 4, and A1 all illustrate this point). Contrary to the findings of Boughton (1991), there is no evidence in any of the decompositions that the recent price weakness is part of an anomalous cycle. This would suggest that the prospects for a sharp recovery in real commodity prices to pre-1980s levels appear somewhat dim. Second, irrespective of the technique used, it is evident that the downward trend has steepened in recent years, implying that the design of stabilization efforts should incorporate this feature of commodity price behavior. Third, while for some of the indices (all-commodities and metals) the results on stationarity are not clearcut, from a practical policy standpoint this distinction may not be relevant. As noted earlier, the success of stabilization funds requires shocks to be temporary. However, if a shock is temporary, but its effects persist for many years, price stabilization may be equally infeasible.

1. Secular developments

a. Economic growth and the demand for commodities

Analyses of commodity market developments have usually focused on the macroeconomic conditions in industrial countries as the principal factor affecting commodity prices. Given the role of many commodities as inputs for production by manufacturing industries, their demand is closely related to the level of industrial economic activity, the major part which takes place in industrial countries. Therefore the trend decline in real commodity prices is, at least, partially related to the marked and secular slowdown in the growth of output in the industrial countries in the early 1970s, with 1973 constituting the end of the strong postwar expansion phase for most countries. ^24/

In addition to the slowdown in growth, a declining intensity of resource use, particularly in industrial countries, has contributed to reduced rates of growth in the demand for commodities. Table 8 presents data for the G-7 countries on the growth of real GDP and consumption for several major metals over two subperiods; the data document a widespread and persistent slowdown across countries and across commodities. In addition, Table 8 reports changes in an intensity of use measure, (i.e., metal consumption per unit of real GDP) which is designed to give an indication of longer-term trends in technological innovation or in structural factors governing resource use. While there are differences between countries and particular metals, the general picture that emerges is consistent with a downward shift in the intensity of metal use between the periods. ^25/

Table 8.

Intensity of Metals Use in Major Industrialized Countries

(Average percentage changes)

^1/I - Intensity of use, i.e., consumption over real GDP.

Table 8.

Intensity of Metals Use in Major Industrialized Countries

(Average percentage changes)

	1962-1973			1974-1991
	Consumption	Real GDP	IU 1/	Consumption	Real GDP	I 1/
Aluminum
United States	9.3	3.5	4.1	-0.1	2.2	-2.4
Japan	20.3	9.5	13.0	3.2	3.9	-0.8
France	7.4	4.9	3.1	3.1	2.4	0.6
Germany	9.6	4.2	6.3	3.1	2.1	0.9
Italy	10.5	5.3	7.7	4.6	2.8	1.6
United Kingdom	5.2	3.1	1.7	-0.5	1.7	-2.2
Canada	8.6	5.3	1.4	3.1	3.0	-0.0
Copper
United States	4.7	3.5	-0.6	0.5	2.2	-1.9
Japan	11.2	9.5	4.2	2.3	3.9	-1.6
France	4.6	4.9	-0.2	1.2	2.4	-1.2
Germany	2.7	4.2	0.2	2.0	2.1	-0.2
Italy	3.4	5.3	0.3	2.8	2.8	-0.1
United Kingdom	0.4	3.1	-5.1	-3.5	1.7	-5.2
Canada	6.1	5.3	-2.5	-0.4	3.0	-3.5
Lead
United States	3.5	3.5	-0.7	-0.3	2.2	-2.5
Japan	6.9	9.5	-1.5	3.3	3.9	-0.7
France	3.1	4.9	0.8	1.0	2.4	-1.4
Germany	3.5	4.2	-0.7	1.2	2.1	-0.9
Italy	9.0	5.3	6.9	1.1	2.8	-1.8
United Kingdom	0.3	3.1	-4.2	-0.1	1.7	-1.8
Canada	3.8	5.3	-1.7	2.7	3.0	-0.5
Zinc
United States	7.1	3.5	2.0	-2.4	2.2	-4.7
Japan	11.4	9.5	2.3	0.7	3.9	-3.2
France	4.0	4.9	1.1	0.6	2.4	-1.8
Germany	3.3	4.2	-0.3	1.6	2.1	-0.5
Italy	7.9	5.3	3.0	2.1	2.8	-0.8
United Kingdom	1.5	3.1	-2.0	-2.3	1.7	-4.0
Canada	7.6	5.3	0.2	0.1	3.0	-2.9
Tin
United States	1.3	3.5	-3.5	-2.1	2.2	-4.3
Japan	8.6	9.5	1.6	-0.2	3.9	-4.0
France	1.7	4.9	-2.3	-1.9	2.4	-4.2
Germany	-0.2	4.2	-0.2	1.4	2.1	-0.8
Italy	4.2	5.3	-1.3	-1.2	2.8	-4.0
United Kingdom	-1.4	3.1	-4.6	-2.7	1.7	-4.4
Canada	4.2	5.3	-2.0	-3.1	3.0	-6.0
Nickel
United States	5.0	3.5	-1.0	-1.2	2.2	-3.4
Japan	17.3	9.5	9.2	3.9	3.9	-0.0
France	6.8	4.9	0.2	2.2	2.4	-0.3
Germany	8.8	4.2	4.2	2.1	2.1	-0.1
Italy	10.3	5.3	7.8	2.8	2.8	-0.1
United Kingdom	2.3	3.1	-4.3	0.6	1.7	-1.2
Canada	8.3	5.3	0.6	4.4	3.0	1.2

^1/I - Intensity of use, i.e., consumption over real GDP.

View Table

Table 8.

Intensity of Metals Use in Major Industrialized Countries

(Average percentage changes)

	1962-1973			1974-1991
	Consumption	Real GDP	IU 1/	Consumption	Real GDP	I 1/
Aluminum
United States	9.3	3.5	4.1	-0.1	2.2	-2.4
Japan	20.3	9.5	13.0	3.2	3.9	-0.8
France	7.4	4.9	3.1	3.1	2.4	0.6
Germany	9.6	4.2	6.3	3.1	2.1	0.9
Italy	10.5	5.3	7.7	4.6	2.8	1.6
United Kingdom	5.2	3.1	1.7	-0.5	1.7	-2.2
Canada	8.6	5.3	1.4	3.1	3.0	-0.0
Copper
United States	4.7	3.5	-0.6	0.5	2.2	-1.9
Japan	11.2	9.5	4.2	2.3	3.9	-1.6
France	4.6	4.9	-0.2	1.2	2.4	-1.2
Germany	2.7	4.2	0.2	2.0	2.1	-0.2
Italy	3.4	5.3	0.3	2.8	2.8	-0.1
United Kingdom	0.4	3.1	-5.1	-3.5	1.7	-5.2
Canada	6.1	5.3	-2.5	-0.4	3.0	-3.5
Lead
United States	3.5	3.5	-0.7	-0.3	2.2	-2.5
Japan	6.9	9.5	-1.5	3.3	3.9	-0.7
France	3.1	4.9	0.8	1.0	2.4	-1.4
Germany	3.5	4.2	-0.7	1.2	2.1	-0.9
Italy	9.0	5.3	6.9	1.1	2.8	-1.8
United Kingdom	0.3	3.1	-4.2	-0.1	1.7	-1.8
Canada	3.8	5.3	-1.7	2.7	3.0	-0.5
Zinc
United States	7.1	3.5	2.0	-2.4	2.2	-4.7
Japan	11.4	9.5	2.3	0.7	3.9	-3.2
France	4.0	4.9	1.1	0.6	2.4	-1.8
Germany	3.3	4.2	-0.3	1.6	2.1	-0.5
Italy	7.9	5.3	3.0	2.1	2.8	-0.8
United Kingdom	1.5	3.1	-2.0	-2.3	1.7	-4.0
Canada	7.6	5.3	0.2	0.1	3.0	-2.9
Tin
United States	1.3	3.5	-3.5	-2.1	2.2	-4.3
Japan	8.6	9.5	1.6	-0.2	3.9	-4.0
France	1.7	4.9	-2.3	-1.9	2.4	-4.2
Germany	-0.2	4.2	-0.2	1.4	2.1	-0.8
Italy	4.2	5.3	-1.3	-1.2	2.8	-4.0
United Kingdom	-1.4	3.1	-4.6	-2.7	1.7	-4.4
Canada	4.2	5.3	-2.0	-3.1	3.0	-6.0
Nickel
United States	5.0	3.5	-1.0	-1.2	2.2	-3.4
Japan	17.3	9.5	9.2	3.9	3.9	-0.0
France	6.8	4.9	0.2	2.2	2.4	-0.3
Germany	8.8	4.2	4.2	2.1	2.1	-0.1
Italy	10.3	5.3	7.8	2.8	2.8	-0.1
United Kingdom	2.3	3.1	-4.3	0.6	1.7	-1.2
Canada	8.3	5.3	0.6	4.4	3.0	1.2

^1/I - Intensity of use, i.e., consumption over real GDP.

Various explanations have been put forward for the trends observed in the intensity of metals use in the industrialized countries. Some of these explanations stress the changing composition of output: the shift from manufacturing to services; the declining share in GDP of gross domestic investment, including structures, which is more metal intensive than other components of expenditure; and, more recently, declines in the defense-related industries. Other explanations focus on the role of resource-saving technologies and the process of materials substitution.

Much less information is available on intensity of metals use outside the industrialized countries. For the former CMEA countries, Dobozi (1990) shows that the period of growing metals consumption and rising intensity of use also came to an end in the mid-1970s as output growth in the former CMEA countries also slowed. However, rather than falling as in the industrial countries, intensity of metals use stagnated or showed little change before declining sharply in the 1980s. As regard developing countries, the rapid pace of industrialization in the Asian region and in parts of Latin America in the 1960s through to the early 1980s has been linked with rising intensity of resource use, and appears to be related to the relatively high levels of gross domestic investment being maintained during this period. The subsequent picture, however, is mixed as external financing difficulties lowered investment and growth rates in many developing countries, particularly in Africa and Latin America.

b. The surge in supply

As shown in Borensztein and Reinhart (1994), supply conditions have played a key role in explaining the weakness in commodity prices, particularly since the mid-1980s. In part, the reasons for the supply response in the 1980s can be found in the debt crisis and the need to adopt adjustment policies that, among other things, aimed at increasing export earnings through an expansion in export volumes. For instance, during 1984-88 the commodity supply index increases at an a 13 percent annual rate, almost three times as fast as the 5 percent annual growth rate of the previous ten years. ^26/

However, there are “structural” persistent trends in technology, information diffusion, and factor productivity that have allowed for sharp and sustained increases in the supply of numerous commodities during the past decade. While there are marked differences between commodities and quantification is difficult, there are a number of important examples of where these type of technology-driven shift factors can be shown to be important.

The diffusion of information about prices, techniques, and marketing opportunities can bring about significant increases in overall commodity supplies through new entrants into international markets. The emergence during the 1980s of first Malaysia and then Indonesia as major cocoa producers is one of the prime examples of this phenomenon. Whereas producers other than these two countries on aggregate increased production by about 15 percent between 1980-81 and 1991-92, Malaysia and Indonesia increased their combined raw cocoa output by 500 percent. During the 1970s, the larger plantations in Malaysia became aware that the rise in real cocoa prices would make production profitable; decisions were made to move into cocoa and the rate of plantings accelerated accordingly. Later, the smaller estates and smallholders followed suit. ^27/ Rising production levels and export earnings in Malaysia did not go unnoticed by the authorities and agricultural companies in Indonesia; in addition, many Indonesians received firsthand experience of cocoa growing by working on Malaysian cocoa plantations. As a result, plantings there grew rapidly through the mid- and late 1980s, and Indonesia rose from being a small producer of specialty “fine” cocoa to producing 167,000 tons of mostly bulk cocoa in 1991-92. While declines in cocoa prices since the mid-1980s have had an effect on other (higher cost) producers, cocoa production remains profitable in Indonesia due to high productivity.

Minerals and metals also provide examples where the accumulation of knowledge and technological progress have led to significant shifts in long-run supply curves. However, the most remarkable examples of information diffusion and technologically-driven productivity growth are found in agriculture. Despite demand growth, particularly due to high rates of population growth in developing countries, the rise in production has been such that real food prices have, with the exception of a brief episode in the mid-1970s, declined significantly in the post-war period. ^28/ While land under cultivation has increased in many parts of the developing world, a sustained increase in yields over time has played a major role in the agricultural supply-demand equation. Cereals provide the premier example, but other crops have also registered significant increases in yields. Advances in agroscience that have led to the development of higher-yield and more disease-resistant varieties, the increased use of fertilizers and pesticides, and the spread of irrigation are the principal factors cited for the increases in agricultural yields (see Chart 7). The improvement in yields was a more or less continuous process in the agricultural sectors of the industrial countries, but in the developing sharp increases came more recently. High-yielding wheat and rice varieties were introduced in a number of developing countries and the rate at which high-yielding varieties were adopted in many cases was rapid.

Chart 7.

Cereal Yields by Region

Developed Countries

View Expanded

Source: FAO Yearbook

The improvements in technology are likely to be “irreversible” in nature and are not limited to a few commodities. These developments have had widespread effects and have led to improving yields for many types of crops, including the beverages (cocoa, coffee, tea), oilseeds, vegetables, and agricultural raw materials such as cotton. It seems likely, particularly in light of recent major advances in biotechnology, that in the future as in the past innovation and information diffusion will continue to play a significant role in increasing agricultural yields still further, possibly extending the observed secular decline in food prices.

There is also little doubt that the agricultural policies of industrial countries have been an important factor in the rapid expansion in world commodity supplies during the 1980s and has contributed to the weakness in prices for many agricultural products. ^29/ National (or transnational in the case of the EC) agricultural policies in the industrial countries typically act to stimulate output and discourage consumption, thereby reducing import needs or increasing supplies available for export. Such government intervention has been a common feature in industrial countries in the post-WWII period; however, a marked deterioration in the agricultural trade environment took place in the 1980s and agricultural trade disputes became increasingly acrimonious, particularly between the United States and the EC. The case of wheat provides an instructive example. Price support under the EC’s Common Agricultural Policy has been so “effective” that wheat supplies surplus to Community requirements have grown over time and have had to be sold on world markets, made possible by the provision of export subsidies. Between 1980-81 and 1991-92., wheat exports from the EC rose by 55 percent to nearly 22 million tons (an increase in market share from 14 percent to 20 percent). In order to counter EC penetration in its traditional markets, the United States introduced the Export Enhancement Program (EEP) in 1985. While trade in wheat has been the most notable source of contention, other grains and commodities (e.g., meat, sugar, oilseeds) have been also affected by similar price support schemes.

2. Cyclical factors

1. Reasons for government intervention

A traditional argument for government assuming much of the commodity price risk is based on microeconomic considerations relating to the welfare and behavior of private agents. ^34/ If the government does not attempt to stabilize commodity prices and supply is price inelastic in the short run, then private agents will bear all the commodity price risk and will face variable and unpredictable income streams over time. ^35/ If the government intervenes to stabilize the prices private agents receive, it is argued, the welfare of risk-averse individuals will increase, as they will enjoy smoother income and consumption streams.

Empirical evidence on savings behavior in developing countries that might shed light on this issue is suggestive but rather limited. However, it has been found that rice farmers in Thailand, for example, smooth their consumption quite successfully both within and between harvest years. ^36/ An analysis of the coffee boom in Kenya of 1976-79 found that as much as 60 percent of boom income went into private savings; farmers were aware that the boom was due to frost damage in Brazil and that their income gains were largely of a windfall nature. ^37/ This is not to say that there were no ill effects in Kenya from the boom (including those emanating from government behavior), but only that on the microeconomic level the ability of private agents in developing countries to absorb some level of commodity price risk and smooth consumption may be substantially greater than many policymakers have assumed or allowed. On the macroeconomic level, the policy implication is that governments may be well advised to reassess their intervention strategies and the responsibility they assume for managing consumption (or more generally absorption) risk at the country level.

A second argument for government intervention rests on the “hysteresis effects” of large temporary destabilizing shocks. A sufficiently large price decline, if passed on entirely to the producer, may prove to have persistent (or permanent) effects on supply. For example, in the case of coffee, new tree plantings may not take place and care of the existing trees may lapse, leaving productive capacity reduced for an extended period of time and, potentially, resulting in loss of market share.

A third reason for intervention is that in some cases, especially for countries with a heavy dependence on minerals, the principal concern is with the fluctuations in government revenues. Often, a fund is established to act as a buffer for government expenditures and financing.

2. Short-term government intervention strategies

For the reasons discussed above, as well as for a variety of other considerations, most governments have chosen to assume a role in attempting to deal with the effects of fluctuations in international commodity prices on the domestic economy. The nature of the intervention has been varied, but the most often-used strategies are outlined below.

3. Government policies for the longer term

The previous sub-sections have discussed policies aimed at smoothing and mitigating the impact of international commodity price variability. But commodity-exporting countries have also had to come to terms with the downward trend in real prices. The earlier analysis suggests that it would be unwise to base policies on the assumption that the weakness in real commodity prices will be quickly overcome. A stronger growth performance over the medium term in the industrial countries would certainly help, as would continuing high rates of growth in the NIEs. Some supply-side factors may ease over time (e.g., recent shocks to metals markets from the former Soviet Union), but other factors (e.g., technological innovation) seem unlikely to be reversed. If such is the case, those countries most seriously affected by permanent or highly persistent commodity price declines must face the necessity of adjusting to the shocks and orienting their macroeconomic and structural policies towards this end.

In assessing adjustment strategies it is useful to examine the dependence of countries on a limited range of export commodities, the extent to which export structures have changed over time, and how primary-product export volumes (and market shares) have evolved in recent years. As Tables 9 and 10 show, in addition to diversifying into manufacturing exports (Table 1), a number of countries in Asia widened the range of primary commodities they exported and sharply expanded primary product export volumes (and market shares). Latin-American countries did not match the Asian performance, while countries in the African region fared poorly in terms of both diversification and volume of exports.

Table 9.

Changing Export Structure: Selected Developing Countries, 1970 and 1990

Source: NCTAD, Handbook of International Trade and Development Statistics, various issues.

^1/“d” is a measure of diversification and “c” is a measure of concentration. Both indices range between zero and 1. The smaller d is, the more diversified exports are. Similarly, the higher c is, the more dependent an economy is on a small range of products.

Table 9.

Changing Export Structure: Selected Developing Countries, 1970 and 1990

		1970		1990
		d 1/	c 2/	d 1/	c 2/
Africa
	Burundi	0.923	0.826	0.970	0.744
	Côte d’Ivoire	0.863	0.422	0.878	0.345
	Kenya	0.813	0.336	0.675	0.239
	Mauritania	0.952	0.864	0.938	0.621
	Senegal	0.793	0.311	0.831	0.251
	Tanzania	0.848	0.255	0.844	0.250
Asia
	Malaysia	0.792	0.371	0.581	0.184
	Pakistan	0.823	0.264	0.845	0.226
	Philippines	0.855	0.324	0.693	0.285
	Sri Lanka	0.935	0.584	0.806	0.279
	Thailand	0.834	0.262	0.598	0.098
Latin America
	Argentina	0.764	0.222	0.650	0.142
	Bolivia	0.891	0.542	0.906	0.321
	Brazil	0.718	0.335	0.529	0.101
	Colombia	0.773	0.622	0.710	0.296
	Mexico	0.584	0.116	0.497	0.311
	ruguay	0.874	0.409	0.754	0.223

Source: NCTAD, Handbook of International Trade and Development Statistics, various issues.

^1/“d” is a measure of diversification and “c” is a measure of concentration. Both indices range between zero and 1. The smaller d is, the more diversified exports are. Similarly, the higher c is, the more dependent an economy is on a small range of products.

View Table

Table 9.

Changing Export Structure: Selected Developing Countries, 1970 and 1990

		1970		1990
		d 1/	c 2/	d 1/	c 2/
Africa
	Burundi	0.923	0.826	0.970	0.744
	Côte d’Ivoire	0.863	0.422	0.878	0.345
	Kenya	0.813	0.336	0.675	0.239
	Mauritania	0.952	0.864	0.938	0.621
	Senegal	0.793	0.311	0.831	0.251
	Tanzania	0.848	0.255	0.844	0.250
Asia
	Malaysia	0.792	0.371	0.581	0.184
	Pakistan	0.823	0.264	0.845	0.226
	Philippines	0.855	0.324	0.693	0.285
	Sri Lanka	0.935	0.584	0.806	0.279
	Thailand	0.834	0.262	0.598	0.098
Latin America
	Argentina	0.764	0.222	0.650	0.142
	Bolivia	0.891	0.542	0.906	0.321
	Brazil	0.718	0.335	0.529	0.101
	Colombia	0.773	0.622	0.710	0.296
	Mexico	0.584	0.116	0.497	0.311
	ruguay	0.874	0.409	0.754	0.223

Source: NCTAD, Handbook of International Trade and Development Statistics, various issues.

^1/“d” is a measure of diversification and “c” is a measure of concentration. Both indices range between zero and 1. The smaller d is, the more diversified exports are. Similarly, the higher c is, the more dependent an economy is on a small range of products.

Table 10.

Value and Volume Indices for Agricultural Exports

(1979-81 = 100)

Source: FAO, Trade Yearbook, various issues

^1/Excluding South Africa.

^2/Market economies excluding Japan.

Table 10.

Value and Volume Indices for Agricultural Exports

(1979-81 = 100)

	1970	1975	1980	1985	1991
	Value Indices
N. America	19	57	104	76	105
W. Europe	19	53	105	96	175
Africa 1/	42	66	105	87	94
Latin America	26	58	104	101	104
East Asia 2/	23	52	105	104	158
	Volume Indices
N. America	46	65	103	80	95
W. Europe	54	73	99	129	150
Africa 1/	130	112	101	94	106
Latin America	79	80	92	127	127
East Asia 2/	62	74	102	134	168

Source: FAO, Trade Yearbook, various issues

^1/Excluding South Africa.

^2/Market economies excluding Japan.

View Table

Table 10.

Value and Volume Indices for Agricultural Exports

(1979-81 = 100)

	1970	1975	1980	1985	1991
	Value Indices
N. America	19	57	104	76	105
W. Europe	19	53	105	96	175
Africa 1/	42	66	105	87	94
Latin America	26	58	104	101	104
East Asia 2/	23	52	105	104	158
	Volume Indices
N. America	46	65	103	80	95
W. Europe	54	73	99	129	150
Africa 1/	130	112	101	94	106
Latin America	79	80	92	127	127
East Asia 2/	62	74	102	134	168

Source: FAO, Trade Yearbook, various issues

^1/Excluding South Africa.

^2/Market economies excluding Japan.