Chile: Selected Issues
Author:
International Monetary Fund
Search for other papers by International Monetary Fund in
Current site
Google Scholar
Close

This paper presents four studies on selected issues of the Chilean economy. The new framework for fiscal policy represented by the authorities' target of a central government structural surplus of 1 percent of GDP is also discussed. This study examines whether the variance of the cyclical component of output in Chile and 11 other countries increased during the 1990s, a period during which the variance of inflation declined. The alternative measures of potential output for Chile are also estimated.

Abstract

This paper presents four studies on selected issues of the Chilean economy. The new framework for fiscal policy represented by the authorities' target of a central government structural surplus of 1 percent of GDP is also discussed. This study examines whether the variance of the cyclical component of output in Chile and 11 other countries increased during the 1990s, a period during which the variance of inflation declined. The alternative measures of potential output for Chile are also estimated.

V. Potential Output Growth in Chile, during 1986–2000, and in the Future78

A. Introduction

192. Over the last 15 years, the Chilean economy has experienced an extraordinary expansion. During 1986–2000 the average per capita growth was close to 7 percent, a rate that is significantly higher than seen in any country in Latin America and most Asian economies. This growth “miracle” has been attributed to a number of factors: “right” policies, such as fiscal prudence, a monetary aimed at a low and stable inflation, market oriented structural policies aimed at maintaining property rights, or, in the case of the 1990s, the return of democratic institutions, and possibly a recovery of copper prices.79

193. The purpose of this chapter is to present alternative estimates of potential output growth over the last 15 years and to discuss different scenarios for future potential growth. Estimates of potential output growth are relevant for the formulation of monetary and fiscal policies as they provide an idea of the sustainable, noninflationary pace of output growth. The chapter applies various methods to estimate potential output: Hodrick-Prescott (H-P) filter, the production function method, and econometric methods using a state-space model (or, a “local linear trend” model). The special nature of the mining industry (based on the extraction of a nonrenewable resource) is recognized by splitting the output series in two components, and, in the production function approach, abstracting from the value of resource extraction (the “ground rent”).

194. While average output growth differs little across methods, the profile of potential output growth differs, however, significantly: the production function approach finds nearly no slowdown in potential output growth, in contrast to the state-space model, which finds a marked slowdown, nearly in line with the reduction in actual output growth. Consequently, the estimates of output gaps for 1999 varies significantly, from about 2¾ percent of GDP (the local linear trend model) to a gap of about 5¾ percent of GDP (the production function approach). The large variation in the output gap estimates suggests that the different methods should be applied with caution when used in policy formulation.

195. The chapter also attempts to look ahead and to shed some light on the question of potential output growth during the coming decade, presenting a growth accounting framework based on the production function approach. Two alternative scenarios for future output growth are also presented: one that assumes no increase in the participation rate and a deceleration in the growth of TFP and resulting in 5 percent potential growth rate per annum; and one more optimistic scenario, assuming growing participation rate and a small recovery in TFP growth, resulting in a growth rate of 6 percent per annum. In both scenarios, structural unemployment rate is assumed to remain unchanged and the capital-output ratio to gradually increase in the coming decade, reflecting the perception of above-average rate of return on investment in Chile.

B. Why do We Estimate “Potential Output”?

196. It has for very long been recognized that economic output is subject to temporary or cyclical variations, although explanations on why such variations exist differ from pure supply side theories to more traditional theories based on the stock adjustment or market imperfections. It is difficult, however, to determine the length or amplitude of the cycles. Long cyclical developments, particularly boom periods, are often mistaken for a permanent change in the potential growth rate or the level of output.

197. The distinction between cyclical or temporary effects and permanent effects on the level of economic activity has wide-reaching implications both for fiscal and monetary policies. The implications for fiscal policies are both descriptive and normative:

  • The descriptive implications stem mainly from the cyclical sensitivity of tax revenue, but also reflect, to a lesser extent, the cyclical sensitivity of expenditure, primarily unemployment benefits. The extraction of the cyclical component of output thus provides an estimate of the fiscal balance in the medium term or, the so-called, “structural balance.”

  • The normative implications arise if there is a desire to conduct counter-cyclical fiscal policies as the structural fiscal balance provides—together with other corrections, such as the removal of “one-off” factors—a proxy for discretionary fiscal policy action. Similarly, the structural balance can provide a benchmark for the conduct of medium-term fiscal policies, as is the case in Chile where a structural surplus of 1 percent of GDP is targeted on the central government accounts, assuming among other things a copper price at a “reference level.”

198. Monetary policy is less directly, but not less importantly, influenced by potential output estimates. Potential output estimates have descriptive implications as it provides an important element in the estimation of the “neutral” policy interest rate or, alternatively, the neutral rate of monetary expansion.80 Normative implications follow from the desire to conduct an interest policy that does not have a procyclical impact, or—in the case of monetary targeting—setting a monetary growth rate consistent with low inflation.

C. What Is So Special About Mining?

199. There are good arguments in favor of treating nonrenewable resource extraction, such as mining, differently than other sources of production.

200. Firstly, value added in resource-dependent sectors tends to be less directly and less rapidly influenced by demand factors than other sectors, reflecting long production lags and, in the case of copper, the relatively large importance of a few large mines. Thus, the lumpiness of mining sector output likely reflects lags in the extraction process as well as speculative accumulation of extracted minerals to take advantage of expected price increases (estimates of mining production is based on actual sales of minerals). The cycle is likely to be nontypical and a model calibrated or developed in large countries with a relatively small mining sector (in relation to total GDP) such as the United States are not necessarily applicable to the Chilean case where mining is responsible for about 10 percent of the economy’s value added (Figures 1 and 2).

Figure 1.
Figure 1.

Mining Value Added-Share of GDP

(percent)

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Figure 2.
Figure 2.

Chile: Annual Growth in MiningValue Added, Other Sectors and GDP, 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.

201. Secondly, the fact that mining is based on the extraction of nonrenewable resource implies that its production will taper off at some time in the future as the resource depletes. Clearly; the speed of depletion is very uncertain as it depends on discoveries of new deposits and technological progress but the future reduction of the production has implications for the sustainability of today’s output level and growth rate. Thus, sustainable national consumption is less than otherwise implied from GDP and its growth rate, a fact that may have implications for medium-term fiscal targets.

202. The nature of mining has therefore important implications for the estimate of potential output. While an increase in nonresource-based production in the current period has normally no implications for production in future periods, an increase in mining production in the current period reduces production in future periods, unless matched by new discoveries or a permanent reduction in extraction costs. Thus, potential output in the mining sector can only be defined over the whole economic life of the resource.

D. Estimating Mining and Nonmining Potential Output

203. This section presents various estimates of potential output based on an explicit recognition of the special nature of mining production. The estimates presented here can be broadly divided into two groups: univariate approaches and estimates based on the production function approach.81

Univariate methods

204. Univariate methods try to decompose the output series in a trend (or permanent) component and a cyclical (or transitory) component. The most popular method was proposed by Hodrick and Prescott in the early 1980s and generates a stochastic output trend (y*) by minimizing a combination of the gap between actual output (y) and trend output, subject to a penalty that constrains the rate of change in the stochastic trend output. The coefficient λ determines the degree of smoothness of the trend:

M i n y * Σ t ( y t y t * ) 2 + λ Σ t [ ( y t + 1 * 2 y t * + y t 1 * ) ] 2 , ( 15 )

205. The usefulness of the Hodrick-Prescott (HP) filter is, however, limited by the lack of a very good method in finding the appropriate smoothing parameter λ—the “standard” parameters values are based on U.S. experience and might not be appropriate for other economies—and the fact that the most recent data tend to influence the estimate of current trends disproportionally (“end-sample” bias). To deal with the former problem, an “optimal” λ can be calculated by conducting grid search in line with Coe and McDermott (1997).

206. The trend components of value added in mining and nonmining sectors was estimated both using the “standard” λ of 1600 for quarterly data and the “optimal” λ (Figures 37). It is interesting to note that the trend growth rate of value added of the mining and nonmining sectors has diverged in the 1990s: while nonmining growth seem to have moderated in the latter half of the 1990s, mining output has been accelerating over most of the decade (Figures 3, 4, and 6).

Figure 3.
Figure 3.

Chile: Mining Value Added and H-P Trend (λ=1600), 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.
Figure 4.
Figure 4.

Chile: Nonmining Value Added and H-P filter Trend (λ=1600), 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.
Figure 5.
Figure 5.

Chile: GDP and H-P Trend ((λ=1600), 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.
Figure 6.
Figure 6.

Chile: GDP and Nonmining Value Added (Trend Growth and Gap), 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.
Figure 7.
Figure 7.

Chile: Nonmining Value Added, λ = Optimal versus 1600

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Benco Central de Chile and staff estimates.

207. As can be seen in Figure 7, the trend was much more curved trend when the “optimal” λ was used, most likely reflecting the larger importance of supply side shocks (less of the movement in the series is considered as temporary or cyclical). The optimal λ was at 724, less than half the “standard” λ While the parameter choice does not seem to matter a lot for the period up until the Asia crisis in 1997, the growth rate of the two series is significantly different in the recent years, when trend growth rate of nonmining value added differs by about 2 percentage points. Irrespectively of which of the two parameters is used in the estimation, the estimated trend growth rate falls quite dramatically, and much more than the common estimates of current potential growth rates (e.g., using the production function approach). This probably reflects the problems attached to the use of the H-P filters after a sharp recession as it is heavily influenced by the most recent observations, whether the standard parameters or “optimal” parameters are used.

208. An alternative approach to estimating trend output was provided by Clark (1987), proposing to estimate the following model in state-space format:

y t = y t * + x t ( 16 )
y t * = y t 1 * + g t 1 + ε t , ( 17 )
g t = g t 1 + γ t , ( 18 )
x t = a 1 x t 1 + a 2 x t 2 + τ t , ( 19 )

209. Where yt is real GDP, y* is a trend, and xt is the cyclical component of output. The error terms εt,γt, and τt are assumed to be mutually independent and normally distributed with zero mean. The stochastic trend is modeled as a random walk with a drift term gt, which allows for permanent shocks to the rate of growth of potential output.

210. The two unobserved components y* and xt, the local linear trend, the coefficients in equation 5 (describing the cyclical developments), and the variance of the error terms can be estimated by using a “Kalman filter” technique, a recursive procedure computing mean squared error estimate of these parameters and unobserved components. Although, a discount factor is used to give recent observations a higher weight than distant observations, the Kalman filter technique does not suffer from end point bias, as is the case of the H-P filter. See, for example, Nadal-De Simone (2000) for a description of the Kalman filter technique and the local linear trend model by Clark (1987).

211. The results for the mining and nonmining sectors are very much as expected (Tables 12 and Figure 8): the mining sectors output does not seem to have a significant cyclical component, in contrast to the nonmining sector were there is a significant first order cyclical parameter (a1). While mining sector output seem to be well described by a model close to a random walk, the nonmining sector seem to be dominated by the cyclical factor, with only small permanent shocks to the potential growth rate or potential output level (σɛ and σγ are not significantly different from zero).

Table 1.

Chile: Mining, Estimation Results: “Clark” Model

Likelihood value: 78.988717

article image

Denotes 5 percent signficance level (at least).

Sources: Central Bank of Chile and Fund staff estimates.
Table 2.

Nonmining, Estimation Results: “Clark” Model

Likelihood value: 78.988717

article image

Denotes 5 percent signficance level (at least).

Sources: Central Bank of Chile and Fund staff estimates.
Figure 8.
Figure 8.

Chile: Local Linear Trend Model: Cyclical Components, 1990-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.

212. The state-space model suggests that the slowdown resulted in a much less dramatic increase in the output gap than found in the H-P filter estimates (and the production function estimates presented below) resulting in a “current” output gap of close to ½ percentage point of GDP (Figures 910). The state-space estimates of potential output growth are, however, highly volatile, particularly in the mining industry. Even in the case of the nonmining sector, where the process appears be dominated by cyclical developments, the estimated potential output growth estimate seems to closely follow actual output growth.

Figure 9.
Figure 9.

Chile: Nonmining Value Added and Local Linear Trend, 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.
Figure 10.
Figure 10.

Chile: Nonmining Value Added and Local Linear Trend , 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.

The production function approach

213. The production function approach is not readily applicable to a resource-based economy such as Chile. The existence of a significant mining industry poses problems of both analytical and empirical nature. The main analytical issue is linked to the nonrenewable nature of mining output. Thus, a significant share of total mining sector sales reflects a “ground rent” or the return on fixed capital invested in the mining sector, in excess of alternative investments with the same perceived risk. The main empirical problem is the lack of long series on investment and employment in the mining industry (the data starts in 1986) and uncertainty attached to the size of the nonrenewable reserves and the cost of extraction.

214. One way to deal with this issue would be to take explicitly into account the potential output from mining over the expected economic life of the resource. As mentioned above, this would imply the estimate of a multiperiod production function where output in any given period would depend on extraction rates in other periods, the resource size and on the investment in the mining sector, which determines the maximum extraction potential in any given period. The estimation of such a multiperiod function goes beyond the scope of this chapter as it would require an in-depth study of the size of the different important nonrenewable resources, the link between capital investment and maximum extraction capacity, and the cost structure of different mines.

215. The alternative approach taken here is to focus on the potential for nonmining use of the input factors, and thus estimate a “normalized” potential output. It should be noted that this output series excludes the ground rent component of mining production; thus, in order to estimate the size of sustainable consumption, an estimate of the permanent income from the copper resource would have to be added to the “normalized” potential output. In practical terms, an estimate of “normalized” potential output, can be achieved by assuming that mining sector value added excluding the ground rent is generated by the same aggregate production function as the rest of the economy:

x m = a + b ( k k r ) + c ( l * l r ) + ε 1 ( 20 )
y r = a + b ( k r ) + c ( l r ) + ε 2 ( 21 )

where xm denotes “normalized” mining sector output; yr is value added of the rest of the economy; l* is “full-employment” labor input; and k capital is the economy’s capital stock. Subscript m denotes the mining sector and r the rest of the economy (and no subscript the whole economy). All variables are in natural logarithms and vary over time (there are no constants apart from the parameters b and c). Potential output is here defined as the sum of “normalized” mining sector output and the value added of the rest of the economy.

216. Accordingly, potential output was estimated in two steps:

  • In the first step, nonmining total factor productivity a was estimated by subtracting b(kr)+ c (lr)—the weighted sum of capital (excluding residential capital) and actual employment (adjusted for hours worked in Santiago82) in the nonmining sector—from total nonmining value added yr (excluding rents and imputed services for owner-occupied housing). The weights used were the standard Cobb-Douglas coefficients (⅓ for capital and ⅔ for employment), which are also in line with coefficients estimated by using error-correction methods on the input factors.83 The resulting estimate of a was then smoothed by an H-P filter (Figure 11).84

  • In the second step, an estimate of potential output was derived by substituting the total economy input factors into the common production function, using the total factor product estimate from the first step: the weighted sum of the total economy potential labor input l* and capital stock k was added to the (smoothed) total factor productivity a. An estimate of rents and imputed services for owner-occupied housing was added separately.

Figure 11.
Figure 11.

Chile: Factor Inputs and Total Factor Productivity, 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.

217. The level of capital stock was calculated by using nonresidential capital stock data from Jadresic (1992) until 1986, and applying the perpetual inventory method with an annual depreciation rate of 5 percent for all the remaining years85 (Figure 11). Due to the absence of reliable information, the share of the capital stock in the mining industry was assumed to be 2.5 percent of total capital stock in 1986, substantially higher than the ½ percent that has been invested in the sector during the last 15 years. Finally, potential labor input l* was estimated using a “smoothed” participation rate and unemployment rate as depicted in Figure 12.86

Figure 12.
Figure 12.

Chile: Trend Unemployment and Participation Rate, 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.

218. The production function approach results in a quite large output gap for the recent years, reflecting a potential output series that grows steadily and does not appear to “suffer” from end-period bias. This may be a bit surprising since the production function approach is a combination of series that has been smoothed by an H-P filter (the participations rate, the unemployment rate, and total factor productivity) and an “unfiltered” capital stock series. This apparent lack of sensitivity to the recent slowdown stems mainly from the fact that potential employment was constructed using a relatively constant structural unemployment rate (the unemployment rate was smoothed by an H-P filter with lambda at 100), and participation rate (Figure 12); by contrast the application of the H-P filter directly on employment would have yielded very similar results to the H-P filter estimations described earlier.

E. How Fast Can Chile Grow in The Coming Decade?

219. The key for assessing the prospects for future output growth in Chile lies with the correct interpretation of the downturn in the late 1990s. Was it the result of a temporary contraction of demand or does it reflect more permanent factors, such as a change in foreign investors perception of the sustainable rate of return in Chile? The sharp drop in copper prices in 1996 represents a quasi-demand side factor as it can have lasting effects on domestic production, but it is unlikely to have reduced potential output for the nonmining sectors over and above the accelerated depreciation of physical and human capital that may happen in any demand induced slump.87

220. The methods employed above do not help much in distinguishing between the different explanations: while the H-P filter technique and the production function approaches tend to interpret most of the sharp fluctuations in output, such as the 1998–99 downturn, as temporary, the local linear trend seem to interpret such sharp movements as largely changes in the potential output level (as well as in its growth rate). A much more complete model of the economy is needed to distinguish between different factors driving the fluctuations in output. The simple production function approach has been extended to include factors such as the terms of trade (Rojas, et. al, 1997) and indicators of external finance (Coeymans, 1999). To some extent these models can be seen as complements to the simple production function approach as they can be interpreted as introducing factors that explains the cyclical developments (indicators of external finance, and to a lesser extent terms of trade) and adjusts the simple aggregate production function by introducing relative prices between sectors (terms of trade). The latter approach can be seen as an attempt to introduce elements of multisector approach to potential output, where the real value of combined output depends on the world prices of the goods from the different sectors.

221. By contrast Barro (1999) presents a set cross-country growth regressions based on: (i) the log-level of per capita GDP (and its squared value); (ii) the ratio of government consumption to GDP; (iii) an index of the rule of law; (iv) democracy index (and its squared value); (v) inflation rate; (vi) years of schooling; (vii) log of the fertility rate; (viii) the ratio of investment to GDP; (ix) growth rate of terms of trade; and (x) a country specific dummy. Reflecting Chile’s high score on most of these indicators and a rapid per capita GDP rate in the past (a factor that weighs heavily in the estimations due to the country specific dummies) the projected per capita growth rate of 1996–2006 amounts to 3 percent, which with a population growth rate at about 1–2 percent would amount to a real GDP growth rate in the range of 4–5 percent per annum.

222. Although insufficient to determine the engine of growth, the production function approach provides elements of a growth accounting framework that proves very useful to organize a discussion about the prospects for potential output growth in the future. The production function approach basically decomposes output growth into two components: (i) the contributions from the input factors, which we can more or less confidently “explain,” and (ii) total factor productivity which is the residual or “unexplained” component.

223. Table 3 presents a growth accounting framework based on the input factors used above in the “production function approach” suggesting a potential growth rate of 5-6 percent in the coming decade, depending on whether the assumptions on the participation rate and total factor productivity are rather pessimistic (no change in the participation rate, the structural unemployment rate, and a slowdown in total factor productivity growth to 2 percent per annum) or more optimistic (1 percent increase participation rate per annum and in the total factor productivity growth rate to 3 percent per annum). Population projections were taken from the World Bank and the participation rate and the trend unemployment rate were assumed unchanged in the next ten years. The investment growth was based on WEO projections up until 2006 and assumed to contribute to a moderate increase in the capital to GDP ratio thereafter.

Table 3.

Chile: Decomposition of Potential Output Growth: 1986-2010

article image
Source: Central Bank of Chile and Fund staff estimates.

224. The growth ranges presented in Table 3 should not be interpreted as projections, but as mere scenarios, illustrating different growth ranges given rather crude assumptions about the input factors. For example, the structural unemployment rate is assumed to remain constant in both the pessimistic and the optimistic scenarios; structural reform in the labor market could reduce this rate, or, on the contrary, the structural unemployment rate could increase if new rigidities are introduced or existing rigidities are intensified (e.g., sharp increase in the real minimum wage). Furthermore, capital accumulation is based on the assumption that investors (mainly foreign) consider the rate of return high enough to merit an increase in the capital-output ratio.

Figure 13.
Figure 13.

Chile: Normalized Potential Output—“Production Function Approach,” 1986-2000

Citation: IMF Staff Country Reports 2001, 120; 10.5089/9781451807585.002.A005

Source: Central Bank of Chile and Fund staff estimates.

References

  • Arellano S. And M. Braun, 1999, “Stock de Recursos de la Economía Chilena,” Cuadernos de Economía, No. 107, (April), pp. 639684.

    • Search Google Scholar
    • Export Citation
  • Barro J., 1999, “Determinants of Economic Growth: Implications of the Global Evidence for Chile, Cuadernos de Economía, No. 107, (April), pp. 443478.

    • Search Google Scholar
    • Export Citation
  • Blanchard O. and Stanley Fischer, 1989, Lecture on Macroeconomics, (Cambridge, Massachusetts: MIT Press).

  • Braun J. and M. Braun, 1999, “Crecimiento Potencial: el Caso de Chile,” Cuadernos de Economía, No. 107.

  • Calvo and Mendoza, “Empirical Puzzles of Chilean Stabilization Policy” in Chile: Recent Policy Lessons and Emerging Challenges, ed. G. Perry and D.M. Leipziger, World Bank Institute Development Studies.

    • Search Google Scholar
    • Export Citation
  • Chumacero, Rómulo A., A Switching Regime Model for Chilean Growth, mimeo, Central Bank of Chile.

  • Chumacero, Rómulo A., J. Quiroz, 1996, “La Tasa Natural de Crecimiento de la Economía Chilena: 1985–1996,” Cuadernos de Economía, pp. 45372.

    • Search Google Scholar
    • Export Citation
  • Clark, P.K., 1987, ‘The Cyclical Component of U.S. Economic Activity,” Quarterly Journal of Economics 102, pp. 797814.

  • Coe, David, and J.C. McDermott, “Does the Gap Model Work in Asia?” IMF Staff Papers, International Monetary Fund.

  • Coeymans, J.E., 2000, Crecimiento a Mediano y Largo Plazo en la Economía Chilena, MIDEPLAN.

  • Coeymans, J.E., 1999a, “Ciclos y Crecimiento Sostenible a Mediano Plazo en la Economía Chilena,” Cuadernos de Economía, (April), No. 107, pp. 54596.

    • Search Google Scholar
    • Export Citation
  • Coeymans, J.E., 1992, “Productividad, Salarios y Empleo en la Economía Chilena: un Enfoque de Oferta Agregada,“ Cuadernos de Economía, No. 87, (August), pp. 22963.

    • Search Google Scholar
    • Export Citation
  • Coeymans, J.E., 1999b, “Determinantes de la Productividad en Chile: 1961-97,” Cuadernos de Economía, (April), No. 107, pp. 597637.

    • Search Google Scholar
    • Export Citation
  • García, Pablo, 2001, “Productividad y Crecimiento en Chile,” mimeo., Central Bank of Chile.

  • Hodrick, R.J. and E.C. Prescott, 1997, “Postwar U.S. Business Cycles: An Empirical Investigation,” Journal of Money, Credit, and Banking, 29, pp. 116.

    • Search Google Scholar
    • Export Citation
  • Jadresic, Esteban, and R. Zahler, “Chile’s Rapid Growth in the 1990s: Good Policies, Good Luck, or Political Change?” IMF Working Paper, WP/00/153 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Jadresic, Esteban, and R. Zahler, G. Sanhueza, 1992, “Producto y Crecimiento Potencial de la Economía Chilena,” Documento de Trabajo, Banco Central de Chile.

    • Search Google Scholar
    • Export Citation
  • Meller, Patricio, R. O’Ryan, and Andres Solimano, 1996, Growth, Equity, and the Environment in Chile: Issues and Evidence, World Development, Vol. 24 (2), pp. 25572.

    • Search Google Scholar
    • Export Citation
  • Meller, Patricio, R. O’Ryan, and Andres Solimano, Un siglo de Economía Política Chilena (1890–1990), Andrés Bello, Santiago.

  • Morandé L., Felipe G., and Rodrigo Vergara, 1997, “Análisis Empírico del Crecimiento en Chile,” Centro de Estudios Públicos, Programa de Postgrado en Economía (Washington: ILADES/Georgetown University).

    • Search Google Scholar
    • Export Citation
  • Rojas, P.R., Eduardo López, Susana Jiménez, Determinantes del Crecimiento y Estimación del Producto Potencial en Chile: el Rol del Comercio Iinternacional, in Morandé et al (1997).

    • Search Google Scholar
    • Export Citation
  • Ramirez, M., 2000, “Public Capital Formation and Labor Productivity Growth in Chile,” Contemporary Economic Policy, 18(2), (April), pp. 15969.

    • Search Google Scholar
    • Export Citation
  • Ramsey, F.P. 1928, “A Mathematical Theory of Saving,” Economic Journal 38, No. 152, pp. 543559.

  • Spilimbergo, Antonio, 1999, “Copper and the Chilean Economy: 1960–98,” IMF Working Paper, WP/99/57 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

ANNEX I I. Summary of Tax System as of May 2001

(All amounts in Chilean pesos)

article image
article image
article image
article image
article image
article image
article image
Sources: Ministry of finance, Directión de Presupuestos, Cálculo de Ingresos Generales de la Nación Correspondiente al Año 2000, Santiago: November 1999 and information provided by the Chilean authorities.

UTA stands for Unidad Tributaria Anual (annual tax unit), and corresponds to 12 times the value of a December’s UTM (Unidad Tributaria Mensual) or monthly tax unit. The UTM is adjusted each month according to the change in average price level in the second-past month. In December 2000 one UTM was worth Ch$27,600 and reflected changes in the price level as per December 31, 1999.

The monthly withheld tax on labor income is computed using the same progressive schedule, but using UTM instead of UTA.

UF stands for Unidad de Fomento, a price reference unit widely used in financial contracts which is adjusted daily. A schedule from the 10th of each month to the 9th of the subsequent month reflects changes in the price level in the previous month. On December 31, 2000, one UF was worth Ch$15,769.92 and reflecled changes in the price level as per December 31, 1999.

ANNEX II Chile: Major Monetary and Banking Measures in 2000–01

February 2000

3

Banks were allowed to issue Deposit Certificates abroad, under the same conditions as those needed to issue bonds.

April 2000

17

The exchange regulations in the CIER were renewed for a period of one year.

20

Chilean companies were authorized to also list their shares in foreign markets in which American Depository Receipts (ADRs) are not traded, as long as they comply with the same conditions that apply for issuing ADRs. At the same time, investment funds for development of companies and real estate investment were authorized to subscribe their quotas in foreign markets.

May 2000

10

Chilean banks were authorized to provide guarantees or finance to foreign financial institutions, but only in operations related to the financing of international trade between third countries.

11

The one-year withholding requirement for foreign investments covered by Chapter XIV of the CIER was eliminated. Along with this measure, it was announced that:

  • Banks and third parties were authorized to trade forward contracts with foreign counterparts involving Chilean currency (either the peso or the inflation-indexed UF).

  • Banks were allowed to hedge credit risk associated with their fixed-income portfolio and commercial loans with residents, using financial derivatives, for both local and foreign currency.

  • The regulation authorizing the issuance abroad of peso- or UF-denominated bonds was revised so that such instruments would be treated, for regulatory purposes, as obligations payable in foreign currency. (One immediate implication was that the tax applied to interest payments on UF-denominated bonds issued abroad was reduced from 35 percent to 4 percent.)

25

Credit unions supervised by the Superintendency of Banks and Financial Institutions were authorized to opt for compliance with the Basle criteria in respect of required capital as an alternative to observing the applicable maximum ratio between their current liabilities and respective paid-in capital and reserves. Thus, credit unions choosing this new alternative must: (i) establish legal reserves of UF 400,000 or more; (ii) have real net worth of not less than 10 percent of their risk-weighted assets and not less than 5 percent of their total assets; (iii) comply with the Central Bank of Chile’s rules on matching; and (iv) obtain the authorization of the Superintendency of Banks and Financial Institutions, which will approve the request if the credit union meets the above requirements and it finds the quality of management and the technical capabilities to be satisfactory overall.

Credit unions that opt for the Basle criteria, comply with the above requirements, establish the required technical reserve, and have due authorization from the Central Bank of Chile were also authorized to issue credit cards and carry out credit card transactions.

August 2000

3

The Clearing House Regulation on checks and other instruments in local currency was amended so that, effective September 13, 2000, the time required to clear checks from other financial centers will be reduced by one day; the funds will therefore be available to users on the fourth day after the check is deposited.

3

Banks were authorized to approve the opening of demand accounts in foreign currency. These accounts will earn no interest and will be nonadjustable.

To reduce the costs of issuing and processing letters of credit and contribute to the current trend toward paperless issues of all types of financial instruments, banks and finance companies were authorized to issue letters of credit in book-entry form, in accordance with the provisions of Law 18.876 on the Organization and Operation of Private Deposit and Securities Custody Institutions. It was specified that if the book entry is made before the letters of credit are placed in circulation, temporary certificates must be issued in physical form, containing the same terms as the letters of credit that would have had to be issued for purposes of the respective mortgage loan.

10

The ceiling on the overdrafts that banks can allow on bank current accounts was eliminated. This could not exceed the equivalent of UF 30 for each customer and was applicable to banking institutions which, in the process of classification of their loan portfolios, had not been rated one or more times in succession in category I by the Superintendency of Banks and Financial Institutions, and for as long as they retain that status.

28

It was decided to reduce the monetary policy rate by 50 basis points, from UF+5.5 percent to of UF+5.0 percent annually. In addition, the liquidity line of credit tranches were adjusted by the same amount, setting the first tranche at the key rate while the second and third tranches were set at UF+7.00 percent and UF+9.00 percent, respectively. The liquidity deposit rate was set at UF+4.00 percent.

September 2000

14

To maximize development of the financial market and continue completing the market, the Board of the Central Bank of Chile made the following changes in the regulations governing savings, deposit-taking and intermediation, and financial supervision, contained in the Compendium of Financial Regulations:

  • Banks and finance companies were authorized to debit—at the request of the holders of term savings accounts, deferred-withdrawal term savings accounts, and demand savings accounts—said holders’ respective accounts with the amounts of life insurance and/or disability premiums they owe an insurance company. These debits will not be considered withdrawals for purposes of determining whether the holder is entitled to adjustments.

  • Banks and finance companies established in Chile were authorized to carry out—among themselves or with any other person domiciled and resident in the country—short sales operations, i.e., operations in which instruments with a restitution obligation are bought or sold, it being further established that the instruments involved in short sales must be instruments issued by the Central Bank of Chile in the context of short sales operations and bonds and letters of credit issued by banks or finance companies, payable in local currency.

  • It was established that the liabilities contracted by banks and finance companies in short sales of the above-mentioned Central Bank of Chile instruments will be exempt from the monetary reserve requirement (encaje monetario).

  • The regulations on interest rate matching, allowing financial institutions to offset their asset positions in financial investments not included in the “permanent portfolio”(“trading” investments) with the corresponding liabilities, were eased.

  • Banks and finance companies were audiorized to engage in derivatives operations involving fixed-income instruments issued by the Central Bank of Chile in the context of open market operations as well as bonds and letters of credit issued by banks and finance companies established in Chile, payable in local currency.

14

In addition to the assets in their portfolio of loans or previously authorized investments, banks and finance companies were authorized to sell or transfer the following to securitization companies and securitized credit investment funds:

  • Nonendorsable mortgage loans, issued by themselves or by other financial institutions, with the exception of mortgage loans granted through the issuance of letters of credit as mentioned in Title XIII of the General Banking Law;

  • Loans in their commercial loan portfolio;

  • Outstanding balances resulting from the sale of real property received in payment of past-due debts or acquired in a judicial auction in the same circumstances, in accordance with the provisions of Article 84(5) of the General Banking Law.

It was further established that as in the case of investments in securities issued by the Central Bank of Chile, the requirement of at least one annual sale or transfer will not apply to loans in the commercial loan portfolio or to the above-mentioned outstanding balances.

To promote modernization of the country’s payment systems, the Board of the Central Bank of Chile decided to adopt a plan of action to bring the country’s payment systems into conformity with the highest international standards in this area.

The Board of the Central Bank of Chile agreed the following:

  • To approve a work program designed to promote rapid modernization of the payment systems. To that end, the central bank will work closely and will form one or more working groups with the authorities of the Superintendency of Banks and Financial Institutions and other official entities, as well as with the Association of Banks and Financial Institutions.

  • To define, with advisory assistance from the above-mentioned working groups, the minimum security, transparency, competency, risk-management, and other requirements that large-value payment mechanisms should satisfy. In addition, to ensure that these systems meet the standards of the “Committee on Payment and Settlement Systems” of the Bank for International Settlements (BIS) in Basle.

  • To place in operation in the central bank, beginning in March 2001, an electronic system for bidding on notes and other open market operations;

  • In the short term, to amend the Clearing House Regulation on checks and other instruments in local currency, so as to expedite the process and announce clearing balances on the same day that the central bank makes the final settlement.

  • In the short term, to make progress in creating a system of financial institution current accounts with the central bank, to facilitate direct access to book balances during banking hours.

  • In the future, to adopt the measures, actions, and regulations necessary for a real-time, online interbank payment system known internationally as the RTGS (Real-Time Gross Settlement System), in which payments can be settled on a gross basis and in real time, in current accounts with the central bank.

October 2000

5

The Central Bank of Chile issued a favorable prior opinion on the regulations proposed by the Superintendency of Banks and Financial Institutions, which amend the rules on investments and credit operations carried out by banks abroad or across borders, with a view to facilitating the process of bank internationalization.

For purposes of assigning country risk to credit guarantors, the risk rating that the guarantor bank must have is reduced from risk category one to investment grade. In turn, sureties, guarantors, joint and several debtors, and issuers of stand-by letters of credit are considered credit guarantors. This rule also applies to operations backed by insurance or credit derivatives.

  • Securities quoted on official stock exchanges of countries rated at least in category BB- or its equivalent are exempt from the obligation of establishing country risk provisions. Previously, category AA was required as a minimum.

  • No country-risk provision is required for loans granted to banks with a residual term to maturity of up to 180 days, nor for operations conforming to certain established definitions, provided that the Board of the respective financial institution has approved the specific types or groups of operations as well as the countries involved.

  • Operations subject to the ceiling of 70 percent of real net worth include loans in foreign currency to enterprises listed on exchanges in countries with a risk rating of at least BB- or its equivalent. These operations are exempt from the 20 percent or 30 percent margin of real net worth.

  • Short-term investment instruments without a risk rating may be given the rating of the long-term instruments of the same issuer, with some additional safeguards.

  • The risk-rating requirement for short-term investment instruments is lowered by one grade to bring the risk of these instruments into line with long-term instruments.

  • Investment alternatives are expanded to include financial investments with no short-or long-term risk rating and whose issuers are located in countries rated at least in category 4 according to the provisions established by the Superintendency of Banks and Financial Institutions. Banks with a Basle indicator of less than 10 percent can make such investments up to an amount equivalent to 10 percent of their real net worth, while for those with a higher indicator this limit is 15 percent of their real net worth.

  • Investment alternatives are expanded to include investments in securities issued or guaranteed by international organizations of which Chile is a member.

  • Financial investments in securities with no risk rating, issued or guaranteed by governments or central banks, will be given the international risk rating of the country in which the issuer is located.

  • Financial investments in structured notes must be issued by investment banks with an international risk rating of AA- or better, and the return on such notes must be pegged to a fixed-income sovereign or corporate instrument with a rating of BB- or better.

January 2001

18

AFPs were authorized to invest up to 5 percent of the type 1 fund in mutual fund shares.

The divisions between the various types of investment fund shares were eliminated, and AFPs were authorized to invest up to 25 percent of the type 1 fund in shares of investment funds and mutual funds.

In addition, the investment ceilings were increased for said type 1 fund, from 37 percent to 40 percent for shares of open corporations and from 10 percent to 40 percent for shares of open real estate corporations.

At its monthly monetary policy meeting, the central bank board voted to lower the policy-related interest rate by 25 basis points, from UF+5.00 percent to UF+4.75 percent. The liquidity line of credit tranches were also adjusted by 25 basis points, to UF+4.75 percent, UF+6.75 percent, and UF+8.75 percent. The liquidity deposit rate was reduced to UF+3.75 percent.

25

As part of payment systems modernization, clearing house regulations on checks, operations between finance companies, and payment obligations resulting from ATM transactions were included in the Compendium of Financial Regulations. The regulations of these clearing houses govern the exchange, clearing, and collection of the respective operations.

February 2001

12

Financial institutions were authorized to make unrestricted transfers (by endorsement) of Adjustable-Rate Notes of the General Treasury of the Republic issued pursuant to Law 19.568 and its Regulations, contained in Supreme Decree 946 of the Ministry of Finance, published in the Official Gazette of September 28, 2000.

20

Institutions showing a current account deficit are authorized, pursuant to Chapter III.H.2, Title II, No. 8 of the Compendium of Financial Regulations, to request funds from the Monetary Operations Office between 4:50 p.m. and 5:10 p.m. to cover said deficit, chargeable to the liquidity line of credit, the cost of which will be the third tranche rate. Alternatively, institutions showing a current account deficit may apply to the same office during the same hours to make counter purchases of Central Bank of Chile notes with repurchase agreement to cover said deficit, the rate of interest for these operations being the rate of the second liquidity line of credit tranche.

20

At its monthly monetary policy meeting, the central bank board voted to lower the policy-related interest rate by 25 basis points, from UF+4.75 percent to UF+4.50 percent. The liquidity line of credit tranches were also adjusted by 25 basis points, to UF+4.50 percent, UF+6.50 percent, and UF+8.50 percent. The liquidity deposit rate was reduced to UF+3.50 percent.

March 2001

02

Meeting in special session, the central bank board voted to lower the policy-related interest rate by 50 basis points, from UF+4.50 percent to UF+4.00 percent. The liquidity line of credit tranches were also adjusted by 50 basis points, to UF+4.00 percent, UF+6.00 percent, and UF+8.00 percent. The liquidity deposit rate was reduced to UF+3.00 percent.

15

Credit unions having opted for the capital and other requirements set forth in Chapter III.C.2, numeral 2.1, of the Compendium of Financial Regulations and, in addition, having obtained the requisite approval of the Superintendency of Banks and Financial Institutions, are authorized to open and maintain the housing savings accounts referred to in Chapter III.E.3 of said Compendium.

April 2001

10

At its monthly monetary policy meeting, the central bank board voted to lower the policy-related interest rate by 25 basis points, from UF+4.00 percent to UF+3.75 percent. The liquidity line of credit tranches were also adjusted by 25 basis points, to UF+3.75 percent, UF+5.75 percent, and UF+7.75 percent. The liquidity deposit rate was reduced to UF+2.75 percent.

June 2001

12

At its monthly monetary policy meeting, the central bank board voted to lower the policy-related interest rate by 25 basis points, from UF+3.75 percent to UF+3.50 percent. The liquidity line of credit tranches were also adjusted by 25 basis points, to UF+3.50 percent, UF+5.50 percent, and UF+7.50 percent. The liquidity deposit rate was reduced to UF+2.50 percent.

13

Credits that result from repo sales of financial instruments expressed in, indexed to, or payable in foreign currency, among banks registered in the country do not need to have the same currency denomination than the original instrument any more.

ANNEX III Chile: Major Exchange and Foreign Trade Measures in 2000–20001

February 2000

3

Banks were allowed to issue Deposit Certificates abroad, under the same conditions as those needed to issue bonds.

April 2000

17

The exchange regulations in the CIER were renewed for a period of one year.

20

Chilean companies were authorized to also list their shares in foreign markets in which American Depository Receipts (ADRs) are not traded, as long as they comply with the same conditions that apply for issuing ADRs. At the same time, investment funds for development of companies and real estate investment were authorized to subscribe their quotas in foreign markets.

May 2000

10

Chilean banks were authorized to provide guarantees or finance to foreign financial institutions, but only in operations related to the financing of international trade between third countries.

11

The one-year withholding requirement for foreign investments covered by Chapter XIV of the CIER was eliminated. Along with this measure, it was announced that:

  • Banks and third parties were authorized to trade forward contracts with foreign counterparts involving Chilean currency (either the peso or the inflation-indexed UF).

  • Banks were allowed to hedge credit risk associated with their fixed-income portfolio and commercial loans with residents, using financial derivatives, for both local and foreign currency.

  • The regulation authorizing the issuance abroad of peso- or UF-denominated bonds was revised so that such instruments would be treated, for regulatory purposes, as obligations payable in foreign currency. (One immediate implication was that the tax applied to interest payments on UF-denominated bonds issued abroad was reduced from 35 percent to 4 percent.)

June 2000

1

It was determined that enterprises that have issued ADRs need not increase their capital in order for their securities to be traded and listed on the various exchanges on which they were issued.

October 2000

12

To promote the development of financial intermediation in foreign currency by banking institutions with persons domiciled or resident in Chile, the Central Bank of Chile voted to modify and supplement the regulation on deposits and loans in foreign currency set forth in various chapters of the Compendium of Financial Regulations and the Compendium of International Exchange Regulations, thus taking an import step forward in the process of liberalizing the exchange regulations applicable to banking enterprises established in Chile.

Progress was deemed possible in this area, seeing that banks, in accordance with the regulations established by the Superintendency of Banks and Financial Institutions, must consider the foreign exchange mismatches of their customers as part of credit risk. The experience of working with a floating exchange rate for more than a year has helped banking institutions better understand the effects of exchange rate volatility on their customers’ financial position and solvency.

As a result of these modifications, banking institutions may also engage in the following operations in foreign currency:

1. Take any foreign currency deposits authorized by the regulation in force, including through savings accounts, whether the latter are demand or term accounts with unrestricted or deferred withdrawals.

2. Extend credit in connection with current accounts and authorize overdrafts on such accounts, in foreign currency.

3. Make commercial and mortgage loans in foreign currency by issuing general-purpose letters of credit to persons domiciled and resident in Chile.

4. Participate in syndicated loans in foreign currency, granted to persons domiciled and resident in Chile, with the exception of banking institutions established in the country, as well as purchase them in whole or in part.

5. Purchase loans granted abroad to persons domiciled and resident in Chile, including loans to finance foreign trade and excluding those granted to banking institutions established in the country.

6. Purchase bonds in foreign currency issued by persons domiciled and resident in Chile, including investment (but only for purposes of intermediation) in bonds issued by the State or by State institutions.

7. Issue certificates of deposits, notes, and bonds in foreign currency, as well as in local currency, adjustable based on changes in the UF, to be placed abroad and traded in Chile and/or abroad.

8. Sell or assign certain assets in their foreign currency portfolio to other banking institutions or to persons domiciled or resident abroad.

The central bank board authorized banks to carry out these new operations in foreign currency insofar as their boards have established policies on assessing their debtors’ exchange risk and have notified the Superintendency of Banks and Financial Institutions of the adoption of such policies in advance.

Moreover, as part of the same objective of the Resolution, it was decided to ease the risk rating requirement for overseas issues of banking institutions’ bonds, convertible bonds, junior bonds, and ADRs.

April 2001

16

The Board of Directors of the Central Bank of Chile decided not to renew the country’s exchange restrictions and approved a new Compendium of International Exchange Regulations, totally replacing the previous version. At the same time, it approved a series of amendments to bring the Compendium of Financial Regulations into line with the new Compendium of International Exchange Regulations. With these new regulations, the central bank culminated a gradual process of deregulating the exchange market, providing individuals and businesses with smoother and more efficient access to the benefits of Chile’s financial and commercial integration with the rest of the world. The new exchange regulations enter into force on April 19, 2001.

In essence, the amendments provide that:

  • Foreign exchange operations of a financial nature continue to be channeled through the formal exchange market, the obligation of settling and/or purchasing foreign exchange on the formal exchange market having been eliminated.

  • Prior authorization is no longer required for capital inflows related to foreign loans, investments, capital contributions, bonds, and ADRs, as is true for capital outflows associated with capital returns, dividends, and other earnings related to capital contributions and investments and the prepayment of external loans.

  • Prior authorization is no longer required for the remittance of capital, profits, and other earnings associated with investments of residents abroad.

  • The limits on special external credit prepayment and acceleration clauses were lifted.

  • The minimum risk rating and minimum weighted maturities restrictions applicable to bond issues were abolished.

  • The limits on the currency in which external debt can be issued or contracted were eliminated.

  • The restrictions on ADR issues were lifted.

  • The required reserve for capital from abroad (which had been set at zero) was eliminated.

  • Foreign trade operations can be carried out on the formal or informal exchange market. In the latter case, such operations must be reported directly to the central bank.

The Board also approved a procedures manual and reporting forms to go with the new Compendium of International Exchange Regulations, some of the provisions of which will enter into force on July 1, 2001.

Table 1.

Chile: Aggregate Demand and Supply

article image
Sources: Central Bank of Chile; and Fund staff estimates.

General government.

Weighted by the contribution to domestic expenditure in the previous year.

Goods and nonfactor services.

Table 2.

Chile: Savings and Investment

(As percent of nominal GDP)

article image
Source: Central Bank of Chile, and Fund staff estimates.

Includes changes in stocks.

Includes central bank losses.

Table 3.

Chile: Sectoral Origin of GDP

(At constant 1986 market prices)

article image
Sources: Central Bank of Chile; and Fund staff estimates.

Includes imputed banking charges, import duties, and value-added tax on imports.

Table 4.

Chile: National Accounts at Current Prices

article image
Sources: Central Bank of Chile; and Fund staff estimates.

Goods and nonfactor services.

Table 5.

Chile: National Accounts at Constant (1986) Prices

article image
Sources: Central Bank of Chile; and Fund staff estimates.

Goods and nonfactor services.

Table 6.

Chile: Indicators of Mining Output

article image
Source: National Bureau of Statistics, as reported in the Monthly Bulletin of the Central Bank of Chile

Includes iodine and nitrate.