IN olden days, sovereigns would often seal treaties of friendship by a marriage of their offspring. This hallowed practice comes to mind when Khan and Montiel (1989) suggest that the compact between the International Monetary Fund (IMF) and the World Bank for growth-oriented adjustment programs be consecrated by a marriage of their respective models. The Fund’s model (hereinafter referred to as FM) reflects the monetary approach to the balance of payments, often used in IMF stabilization programs. The Bank uses a Harrod-Domar growth model called RMSM, which, in its simplest form, derives the potential growth rate of output as the ratio between the savings rate and the incremental capital output ratio (ICOR).

While the authors modestly present their integrated model as “a starting point for the development of a generalized framework” (Khan and Montiel (1989, p. 279)) to analyze adjustment with growth, they also claim that this model “can be used to determine a set of demand management policies..., exchange rate policies, structural policies (policies to increase savings and the level and efficiency of investment), and external financing policies” that would achieve given targets for growth, the balance of payments, and inflation (p. 303).

Although there is a considerable body of knowledge concerning policies that are beneficial to adjustment, to growth, or to both, much work on this subject remains to be done. It would seem unlikely a priori that the merger of FM and RMSM could make an important contribution to this task. Analysis of the Khan-Montiel paper confirms this hunch. More specifically, it reveals three weaknesses of the merged model: (1) The effort to cross FM with RMSM incapacitates each from doing its own job. (2) The simplicity that accounted for part of the attraction of the two models is lost in their merger. (3) The merger provides scant rewards in terms of new insights with respect to growth-oriented adjustment.

These three propositions are spelled out in the remainder of this note.

(1) The authors link FM and RMSM by setting the growth rate in the former equal to that produced by the latter. They justify this procedure on the grounds that FM does not contain an equation explaining the growth rate. This is formally correct; as a recent Fund description of FM explains, “real income is treated as though it were exogenous” (International Monetary Fund (1987, p. 13)). But the absence of an equation does not mean that FM treats output as a loose end for which one can plug in just any number, such as the outcome of the RMSM equations. For, as the same Fund paper continues, “in the actual formulation of Fund-supported adjustment programs, the implications of policies for both output and the price level are carefully analyzed and, of course, output and inflation targets are major factors in deciding upon the policy package” (p. 13). Two crucial questions arise in connection with many Fund programs: Can the country’s economy respond to improved price stimuli, such as devaluation, with a quick burst of additional output? Would the fiscal and credit measures of the program lead to a contraction of output well below capacity levels? Both questions vanish into thin air if the assumption of full capacity use is imposed on FM.

RMSM also suffers in the merger. In RMSM the flow of foreign savings is exogenous. The merger makes it partly endogenous, and in this way domestic credit creation is found to enhance “foreign savings,” because, by raising prices, it worsens the current account. This line of reasoning is incompatible with RMSM. If a country has spare reserves to finance a current account deficit, this element of foreign savings enters RMSM from the start. If the country does not have the reserves, it cannot allow the payments deficit to develop; that is, the merged model is not applicable.¹

(2) Obvious technical problems—the consequences of which remain to be investigated—surface in the attempt to graft a multiplier-type model such as FM, which approaches an equilibrium value, on an exponential-type model such as RMSM, which grows like compound interest. Some of these problems are set out in the last section of the paper, such as that all adjustments in the model are assumed to take place in one period, without allowance for lags or for a gestation period for additions to the stock of capital. Others are implicit in the text, such as that the conclusions are valid only for small changes in the neighborhood of a starting point involving zero inflation and zero growth (d_y = dP_D = 0) (p. 290)—perhaps not the most likely starting point for countries in need of growth-oriented adjustment.

(3) The paper discusses the effects on output, prices, and the current account of seven changes in policy or parameters, some of them monetary and the others raising the supply of savings or factor productivity. Since there is a good deal of duplication among the seven cases,² a brief discussion of three of them will suffice to appraise the policy findings that are credited to the merger exercise.

One such finding is that “consistent with the standard view..., an expansion in domestic credit will raise prices, increase output, and worsen the balance of payments” (p. 294). As regards output, this is not consistent with the standard view, once one allows for the assumption of the merged model that capacity is fully used. In those circumstances, the standard policy advice would be: don’t increase credit creation, because it will only raise prices and worsen the balance of payments. There is, of course, an easy reconciliation, but the merged model is not needed to find it: as mentioned under (1), if a country can afford an increased current account deficit, it can use the resulting resources to raise its growth rate. The country can mobilize these resources in ways other than, and in preference to, generalized credit creation—for example, by import liberalization.

Second, the effects of a devaluation on the balance of payments and on growth are found to be the same (but for a scale factor) as those of a decrease in domestic credit. Thus, if devaluation improves the balance of payments, it must worsen growth—along the line of causation described above. The authors—one of whom co-authored an analytical paper on the question of whether devaluation is expansionary or contractionary (Lizondo and Montiel (1989)—note that this finding is “model specific” (Khan and Montiel (1989, footnote 21, p. 294))—that is, that it may be attributable to quirks in the model.

Third, increased savings provide additional resources for investment. On the basis of RMSM, one would calculate that these changes would raise the growth rate by the amount of resource transfer divided by ICOR. Not so in the merged model, where, as shown in Figure 4, (p. 297), growth is held back by the absence of a corresponding credit expansion. Hence, prices fall, the balance of payments improves through substitution (while it worsens on account of higher growth), and part of the additional resources end up as foreign assets rather than as an increase in the domestic stock of capital.

Again, what is the policy relevance of this finding? Surely, no government that has managed to increase the supply of savings by cutting government consumption expenditure, inducing an increase in the private savings rate, or negotiating loans abroad will then allow the fruits of these policies to be partly lost for lack of an adequate growth in the money supply. Although RMSM tends to be presented entirely in real terms, its growth formula is obviously predicated on the assumption that the economy will be provided with enough money to move the enlarged real turnover without a growth-inhibiting price squeeze. On the assumption of a constant velocity, this yields the unsurprising policy rule that the money supply will have to increase at the same rate as output.

In conclusion, while the authors’ “merged model” offers intellectually interesting exercises, it adds little to our knowledge on the crucial issue of growth-oriented adjustment