The data – the 1995 intercensal survey of Indonesia –the identification strategy, and the specifications follow closely Duflo . The reader should refer to that article for more details. We first replicate Duflo using a dummy for agricultural employment, rather than log wages as in the original paper, and show that the program-induced exogenous increase in schooling led the affected cohorts to work less in agriculture. We then use a different specification to show that the magnitude of the effect of schooling on agricultural employment is in line with the estimates of Section 5.1.Since Schuh’s famous paper, the effects on the agricultural sector of exchange rates and monetary policy have been a subject of much interest and controversy in agricultural economics. Increasing attention has been paid to the role played by shocks emanating from the monetary and financial sectors of the economy. The magnitude and duration of the effects of these shocks on the agricultural sector is still not resolved. For the U.S., some studies find monetary factors to be important. while others disagree.Part of the difficulty in reconciling the different conclusions available in the literature is that no common theoretical model underlies these studies. However, if monetary policies are to be considered important forces in determining agricultural market conditions, a theoretical framework must be developed in which this proposition can be evaluated. Otherwise, empirical analyses which purport to show significant real effects of exchange rates, intl.ation. etc.,lack the theoretical background against which results can be judged. In this paper, we discuss a model of price and exchange rate dynamics in which there are short-run effects of monetary policy. They take the form of relative price changes which benefit the agricultural sector during expansionary monetary policy regimes and which turn against the sector when money is tight.
These results are based on the exchange rate overshooting model of Dornbusch, in which short-run exchange rate changes in response to money growth can exceed. or overshoot,greenhouse ABS snap clamp their long-run equilibrium values. Unlike the Dornbusch model, in which all goods’ prices are fixed, agricultural prices in our model are assumed to be flexible. and we focus on the importance of this assumption for the agricultural sector. The model is consistent with rational expectations and asset market equilibrium at every point in time, and with the long-run neutrality of money. We adopt a “fix-price, flex-price” fr<lm~work, to use the terms originating with Hicks. Prices of agricultural commodities, because those goods are homogeneous. frequently traded, and storable. are assumed to be flexible and governed by instantaneous commodity arbitrage. Non-agricultural goods. on the other hand, are more often differentiated products, with contracting, less rapidly disseminated information, and imperfect competition as possible causes of less rapid price adjustment. Price adjustment therefore occurs instantaneously in the flex-price agricultural markets, while fix-price, non-agricultural ,I markets respond gradually to changes in aggregate demand. The paper has two main sections. The first half focuses on some of the theoretical background and some empirical evidence on the stickiness of prices. First, a review of the model of agricultural price and exchange rate overshooting is given. Next, we consider the factors affecting the degree of overshooting and present some empirical evidence for the United States. Results are presented for Australia which are also suggestive of different speeds of adjustment between agricultural and non-agricultural prices. The second half of the paper describes some simulation experiments using a structural model of the U.S. agricultural sector. In the interest of conserving space, only a brief overview of the model is given, but Figure 1 gives a representation of the model structure. The paper concludes with a discussion of the implications of the results for agricultural policy and suggestions for further research.The overshooting model was developed by Dornbusch to explain variability in flexible exchange rates. In his model, all goods prices were assumed to be sticky, adjusting less rapidly than the pric12s of assets . This causes short-run exchange rate changes in response to changes in the money supply which are greater than the long-run outcome. A simple example illustrates the application of the concept of overshooting to food prices in a pure exchange economy. Consider two goods markets, say, food and widgets, and assume that there is a currency but the quantity of real balances demanded is perfectly inelastic.
When a doubling of the money supply occurs in the presence of perfect price flexibility, doubling of the food and ” widget prices follows immediately and money is neutral. The doubling of the price level leaves the quantity of real money balances unchanged, and equilibrium quantities of food and widgets are also unaffected. The flexibility of prices is the key. Now, assume that food is a ft.ex-price good, while the price of widgets adjusts slowly over time in response to changes in aggregate demand. With short-run fixity in the price in the widget market, such an adjustment is prevented. If the food price alone doubles after a doubling of the money supply. there is excess demand for goods and excess supply of money balances. The continuing effort of money holders to rid themselves of excess money balances guarantees further food price increases. If the widget price gradually rises over time, initial relative prices can be restored. Thus, what would be observed in this simple world is an overshooting of the food price in response to money growth. with the food price falling gradually back to its long-run equilibrium, while the widget price gradually rises to its long-run level. The longer it takes for the latter change to occur, the longer the food price will be above its eventual level. As long as there are no impediments to the eventual doubling of the widget price. the end result is that of the price flexibility case, with relative prices urtchanged. An important factor is omitted from this simple example, the interest elasticity of money demand. It was assumed tpat under no circumstances would individuals hold real money balances in excess of the initial stock. If the quantity of money demanded responds positively to decreases in the interest rate, however. the above result is not necessary. It is easy to see that, the greater the willingness of individuals to hold extra real balances, the less the effect of money supply changes on the food price in the short run. That is. the more interest elastic is money demand, the less will be the degree of overshooting. However, the change in the interest rate brings capital markets into the .’ picture. Departing from this simple model of a closed, exchange economy, let us introduce a world capital market and currencies. Assume that the home country in question is a small country and assume that uncovered interest parity holds, so that home and rest-of-world nominal interest rates differ only by expected depreciation in the value of the home currency.
The domestic nominal interest rate can only change if there is an expected appreciation or depredation of the currency at the same time. Dornbusch used this setup to show that there would be exchange rate overshooting following a change in money growth, as long as the prices of sticky-price goods had not reached their new long-run equilibrium levels. He assumed that all goods were subject to this gradual price adjustment. Frankel and Hardouvelis observed that it is possible to substitute the prices of commodities for currency prices in Dornbusch’s model, and found that asset market equilibrium conditions applied to the market for storable commodities guarantee the same outcome: nominal commodity prices overshoot their eventual levels in response to changes in money growth rates.Stamoulis et. al. used the Frankel and Hardouvelis assumption that agricultural commodities are flexible-price-goods, but kept the exchange rate in the model, as well. The law of one price was assumed to hold at all times for the agricultural commodity, so that its price was never out of line with rest-of-world prices,snap clamps ABS pvc pipe clip while gradual price adjustment again characterized non-agricultural prices. The model thereby differs from the Dornbusch model only in allowing some goods to have flexible prices. The home country is assumed to be a small country in both agricultural commodity and capital markets. This and the law of one price assumption guarantees that the domestic nominal food price will follow exactly the same path as the value of the home currency, so that it overshoots its long-run equili- ” . brium and that this condition persists as long as adjustment in remaining, sticky-price goods markets lags behind. Comparison of the path of adjustment resulting for flexible prices with the Dornbusch solution reveals that, as the number of flexible-price goods in the economy rises, the degree of overshooting is reduced. Thus, the results provide an intermediate case between complete flexibility of prices and the case of stickiness of all goods’ prices. The appendix to this paper contains a short derivation and description of these results. These results depend on few assumptions. The price of agricultural commodities must be free to adjust, as can the exchange rate, while non-agricultural commodities must be characterized by slower adjustment. The model does not require a violation of rationality on anyone’s part, provided one adopts the view that there are good reasons for the existence of contracts, costly price adjustment, or other factors contributing to the stickiness of prices elsewhere in the economy. In fact, Dornbusch showed that overshooting is consistent with rational expectations at every point in time. Also. there is no reliance on the substitution of other countries’ agricul-tural products for the home country’s exports; the law of one price was assumed to hold. To the extent that it is violated. say. because of price supports limiting the downward flexibility of prices. weak export markets can be expected to add to these relative price changes in the case of tight money. In the fix-price. flex-price framework, the short-run effect of monetary policy thus shows up in relative price changes. An expansionary regime favors the agricultural sector in the short run, since relative prices favor agricultural producers, while tight money has the opposite effect, causing larger and quicker decreases in agricultural prices than in non-agricultural prices.Our version of the overshooting model depends on-the assumption that the economy can be described by two types of goods, flex-price goods such as agricultural corrunodities and financial assets, and fix-price. or sticky price goods, such as many non-agricultural commodities. Evidence concerning the relative stickiness of non-agricultural prices comes mostly from ad hoc regressions in which price indices or their growth rates are linked to money growth rates and possibly some other causal factors such as income growth. Based on quarterly data. the evidence is much stronger in favor of the relative stickiness of non-agricultural prices than it is for actual overshooting of agricultural prices. Money growth does appear to have a greater initial effect on agricultural prices, but the effect is probably not greater than one-far-one. We have not tested for overshooting of the exchange rate variable, but if the exchange rate does overshoot its long-run equilibrium. the same result need not apply to the price of agricultural commodities. This is the case if the law of one price does not hold instantaneously for agricultural commodities.Results given in Stamoulis et. al. for the United States were based on the method used by Barra to construct an anticipated money growth variable. Regression of the growth rate of the U.S. Department of Agriculture’s index of prices received by farmers on current anticpated money growth rates revealed an effect much larger than for the Consumer Price Index or for the nonfood CPI. The lagged dependent variable was found to have a larger and significant coefficient in the latter regressions, consistent with the gradual price adjustment assumption used above. Meanwhile, the coefficient on the lagged dependent variable was small and insignificant for the growth rate of the index of prices received by farmers. These results were consistent with those of Lombra and Mehra, who found that the cumulative effect of money growth is I greatest in the consumer price index for food, but that it has larger initial effects. the less is the processing component in the food price index examined. Depending on data and specifications used for estimation. results of tests of neutrality do appear to vary across studies.