by Bob Burkhardt
March 19, 1979
Converted to HTML August 5, 1998.
1 Introduction (August 5, 1998)
This is a short paper I did as a graduate student in economics at M.I.T. as a preliminary investigation into a possible Ph.D. thesis topic. I had switched from monetary economics and was looking into development economics. Richard Eckaus was my advisor. He had written the economics textbook I used in my first economics class at Harbor College. I was surprised to find he had done some interesting work in monetary economics. On reflection though his investigations were much in keeping with his interest in development economics. One of his papers explained loan rationing as a profit maximizing response to the increasing dangers of default associated with higher interest rates. His analysis seemed much more straight forward and reasonable than a lot of things I'd seen monetary economists come up with. Anyway, I didn't get much beyond the LDC loan topic than what is contained here.
This was written before all the LDCs started defaulting on their loans. At this point, some people were wondering what was driving this unprecedented flow of funds.
A version in PDF format is available.
2 Introduction (March 19, 1979)
In the period 1971-1975, LDCs as a whole experienced a significant increase in their long-term capital inflows from abroad. This increase was primarily due to increased inflow from private sources, the Eurocurrency market providing the most dramatically increased inflow. Concessional credit flows from multi-lateral agencies kept up with this overall increase; however, other official development-assistance flows did not. In what follows, I go over some explanations for these increased flows and suggest some possible empirical investigations which would further clarify the phenomenon.
3 A Model
To aid the outlining of explanations, I introduce a simple model. In it, the world rate of interest is determined by the traditional investment-savings equilibrium condition:
where y is the exogenously given level of world GNP, and as and ai are shift parameters. Increases in as and ai result in increases in savings and investment respectively. Since I assume increases in r increase savings and decrease investment, an increase in as decreases the equilibrium r and an increase in a increases the equilibrium r.
I assume that LDCs are a small part of the world economy; so, the determination of r is exogenous to LDC behavior. Their foreign borrowing, net of grants, T, they receive is:
where y* is LDC GNP, and ab is a shift parameter. B responds negatively to increases in r and positively to increases in the rest of its arguments. An increase in T is assumed to decrease B - T.
4 Explanations
4.1 Increase in Savings by Developed Countries
I start the enumeration of explanations by examining changes which would decrease the level of r LDCs face. The first such change is an increase in saving by developed countries. In the first equation, this would appear as an increase in as. A likely source for such a change would be a shift in the world distribution of income from low to high savers. Such a shift occurred in late 1973 when oil-producing nations started restricting supply, thus shifting world income to their pockets. This would be a very attractive explanation for the increased flows to developing countries except the timing is a bit off: the large inflow to LDCs started in 1971. However, it might be counted as one of the factors sustaining the large inflows in later periods.
A more exotic explanation of increases in savings would be increases in the efficiency of financial intermediation (á la Gurley and Shaw). Such an increase, through say the emergence of the relatively unregulated Eurocurrency market, would increase the realized rate of return to savers given the rate of return on capital and thus increase savings. This explanation is appealing in that it connects the growth of LDC borrowing with the simultaneous growth in Eurocurrency lending. There might be something of an identification problem here though since the Eurocurrency market may have just served as a convenient channel for guiding new flows to LDCs. Serious use of such an explanation would require tightening of Gurley and Shaw's theoretical framework.
Another explanation for a declining world interest rate might be a shift in the marginal efficiency of investment (MEI) schedule (the right side of the first equation) due to a change in output (y) or the other factors subsumed in ai. Attributing interest rate changes to changes in output would lead one to expect the borrowing phenomenon to be cyclical in nature. That is not how it appears. I can't think of a source of secular change beginning in 1971.
A final factor which wouldn't change the world interest rate, but would change that faced by LDCs, is a decrease in the perceived risk of lending to developing countries. A priori, this doesn't seem all that plausible as an explanation.
4.2 LDC-Specific Changes
I now move on to examining how changes specific to LDCs might have caused the increase in financial flows I am examining. The first such change might have been a decrease in the supply of concessional credit or other sources of credit to LDCs. While such an explanation might explain the changes in LDC credit sources, it seems contradicted by the major aspect of the phenomenon under investigation, i.e. the increase in total flows to developing countries. A decrease in one source of credit would seem to be a tightening in credit terms to LDCs. That would lead one to expect a decrease in financial flows to LDCs, ceteris paribus.
LDCs may also have demanded increased flows to facilitate the adjustment of their balance of payments. This is definitely plausible in the post-oil-crisis world. Its plausibility for the period before requires further investigation.
Another source of increase in demand might have been a shift in the marginal efficiency of investment in LDCs. Such a shift might come about through increased export earnings. If prices on exported goods increase in terms of foreign goods, investment in export industries becomes more productive. For an extreme case where exports are produced only by foreign owned capital, we might have borrowing determined by an equation like:
| max | p·f(K) - rK |
| K |
where p is the price of exports, f is a production function and K is borrowing. The optimal solution, K* , is increasing in p.
Also, the increase in wealth caused by such a price change may increase the demand for investment in other sectors. These types of explanations seem reasonable in the case of an actual or expected secular rise in export prices. They also have the appealing feature that in a phenomenon like the commodity boom of the early 70's the behavior of a broad range of countries might be affected.
5 Suggested Empirical Explorations
These ideas point to several types of data gathering and analysis activities. The first thing I'd want to do is extend my time series on the composition and magnitude of real financial flows to LDCs as far back and forward as possible. This information should be broken down by GNP since the aggregate statistics are dominated by large countries too much. This would provide information on possible precedents to the present behavior as well as give me a broader look at the present phenomenon.
A second possible investigation is a cross-sectional look at LDC borrowing behavior. External debt growth over a period could be related to GNP growth, initial ratios of debt to GNP or gross exports, and some measure of changes in export earnings. It might also be useful to use other ways of categorizing LDCs. This work should be preceded by a more rigorous theoretical analysis of the determinants of LDC borrowing. This kind of investigation could be done before and after the oil crisis to see how the phenomenon changes, and before the expansion of Eurocurrency lending to see if changes occurred in borrowing patterns of countries. It might be that export income is the driving force initially being supplanted by balance of payments problems later.
A third type of investigation which might be worthwhile is looking at the behavior of world returns on capital. In the U.S., there was talk of a secular decrease in returns to capital. However, there was suspicion that it might be only a cyclical phenomenon. This exchange appeared in the Brookings Papers a couple years ago.