9:42 AM Jun 7, 1996


by Prof Jan Kregel*

Geneva June (TWN) -- It has become common for economists to analyze the long-term performance of the economy in real terms, while relegating money and financial issues to short-term effects.

The problems caused for developing countries by capital flows have been highlighted over the last twenty years. The basic lesson is that too much (as in the 1970s) is just as bad as too little (as in the 1980s), and what is even worse is passing rapidly from one to the other (the 1990s).

These general lessons have given rise to some particular propositions for developing country borrowing.

The first, in response to the expansion of syndicated bank lending to sovereign countries in the 1970s, is that bank borrowing is to be avoided and it is better to use non debt-creating borrowing.

This seems to be an oxymoron. Nonetheless, the idea has been built up that private portfolio and foreign direct investment (FDI) are preferable to bank lending since they do not have a foreign currency denominated fixed interest charge determined by international interest rates.

Recent experience with the volatility of portfolio flows has led to doubts about the benefits of this type of borrowing relative to FDI which is now considered as the preferred alternative. This impression is supported by the reference to the Asian development process which has been largely free from the problems relating to excessive capital flows.

The idea has thus grown that FDI is the most appropriate form of foreign borrowing because it does not have the fixed interest element of bank debt and it does not have the volatility associated with portfolio investment. This is basically because FDI is considered to be investment in domestic bricks and mortar which cannot easily be moved.

The distinction between portfolio and FDI was initially created in order to distinguish between foreign and domestic control of productive assets, the presumption being that investment for purposes of control will be of a more or less permanent variety. The concern over excessive FDI was a policy concern over its nature: if restrictions are imposed to remedy the balance of payments crisis, will foreign owners simply choose to avoid them by closing down and walking away?

Concerns of this sort were common in Europe in the 1960s. But, they are quite different from the kind of policy concerns of most developing countries today. The official definitions of FDI thus have nothing to do with its permanence or mobility or volatility. Indeed, they seem to ignore the facility with which developed country investors obtain and discharge control of companies through mergers and acquisitions markets. In current financial markets companies seem to have become commodities traded daily in markets for corporate control.

But even if FDI were redefined so as to include only investments in immobile bricks and mortar would that mean that FDI would have a different impact on capital flows and exchange rate management? The answer to this is No.

Recent innovations in financial markets have gone a long way towards eliminating the concept of a permanent investment or the "maturity" of an investment. An investor in a 30-year bond used to be considered a long-term or permanent investor holding a long-term asset until maturity. But financial engineering has shown how a 30-year bond can be re-engineered to produce (a minimum of) 61 zero interest rate bonds with maturities ranging from 6 months to 30 years.

I may be a purchaser of some or all of them, or may buy and sell components to produce an instrument which has any number of different possible cash flows and maturities. To reduce my long-term position, and thus the permanence of my investment, I certainly do not need to sell the original bond.

In addition, the availability of futures and options contracts provide other alternatives which allow me to retain ownership of the bond, but to reduce the permanence of my investment as well as exposure to market risks such as changes in interest rates or exchange rates. If this is the case for a 30-year bond investment, it will not be different for foreign portfolio investment.

And it will also be the case of FDI. I can continue to own the bricks and mortar without retaining the foreign country risks, the foreign exchange risks etc associated with their permanent nature.

Usually the methods available for hedging the risks of any investment, including FDI in bricks and mortar, do not lead the investor to engage in any direct foreign exchange or a capital market transaction. However, they will almost always require the financial intermediary to do so at some point in time. And, these flows will usually be greatest when the uncertainty over the exchange rate or domestic financial conditions are greatest.

It is thus simply not the case that, since it is difficult to find a buyer for a bricks-and-mortar-FDI at short notice, they will not produce the same types of financial flows as portfolio investments. Home country firms keep their balance sheets in their domestic currency, and foreign investments represent currency risks in the same way as any other use of company funds, and they will also be hedged in the same way. Just because the investments are in real productive assets does not mean that the foreign currency translation risks and funding risks will be ignored. To the extent that risks are covered they will produce cross-border flows which will produce pressure on the foreign exchange market or the domestic money market.

A World Bank study (World Bank Discussion Paper, # 228), makes the same point, in a slightly different way, by noting that "long-term flows are often as volatile as short-term flows, and the time it takes for an unexpected shock to a flow to die out is similar across flows". The study concludes that in general it is impossible to differentiate statistically between portfolio and FDI capital flows.

Part of this result is, of course, due to the way FDI is classified and can include a wide range of financial investments which are not in any way related to bricks and mortar investments. But, it also highlights the fact that in the presence of the globalisation of finance, all investments have become liquid and all investment will be hedged to some extent.

There is another aspect of FDI investments which is usually overlooked when it is argued that they are preferable to bank loans because they do not generate interest flows. This is the presumption that since there are no prearranged interest payments, there will be no payments made to foreigners. Yet, foreign direct investors do not invest without the expectation of being repaid, with profit.

For a developing country, FDI is not an unconditional gift. It is a loan made against the expectation of profit earnings and the possibility of repatriation.

Now, from the point of view of the lender of funds, the risks associated with bank syndicated sovereign lending are the lowest, risks of portfolio investment higher, and those associated with FDI the highest. In fact, this is partially related to the fact that these investments are more permanent and thus less easy to hedge. But it is also due to less perfect information, the difficulties associated with operating in foreign cultures and the simple factor noted by both Adam Smith and David Ricardo that investors will have a preference for keeping their investments at home where they can keep an eye on them.

Thus, the risk premia attached to the returns on lending will be the highest for FDI. This means FDI is the most costly method of borrowing capital for a developing country. Indeed, most international companies carrying out FDI have hurdle rates of return in the range of 15-25 percent. It is paradoxical that developing countries are told that FDI is the least risky form of foreign borrowing, when from the point of view of the foreign lender it is considered the most risky. If the risk is borne completely by the foreign lender, then compensation will be expected in terms of an appropriately high rate of return.

This leads to another often overlooked aspect of FDI. As is apparent from FDI statistics, in most countries which have benefitted from foreign investment, the greatest proportion of FDI is comprised of the reinvestment of profits on the FDI.

One of the basic reasons why FDI is considered a more attractive form of foreign borrowing than say bank borrowing, is that it does not produce a fixed or immediate charge on foreign exchange reserves. However, profits nonetheless represent a claim on reserves, for they are recorded as current account outflows balanced by an automatic offsetting entry in the capital account representing the reinvested earnings as an FDI inflow.

In a sense this is the equivalent of capitalising the interest on a loan, and simply shifts the claim on foreign exchange reserves into the future. While the recipient country may consider these reinvestment flows as the equivalent of FDI, the investor may consider these flows as a delayed return on the original investment, and they need not be invested in productive enterprise, but may easily be invested by the foreign-owned subsidiary in liquid domestic financial assets. Thus, although they are recorded as FDI flows, they are the equivalent of short-term portfolio investments in both fact and intention. This can make it very difficult to assess the real condition of a country's balance of payments, for they can create the bunching of the repatriation of earnings, creating disruption in the foreign exchange market.

True, BOP conditions may be further complicated if FDI flows finance investments in productive facilities requiring a large proportion of imported specialised capital goods and specialised semi-finished goods for domestic assembly. While these imports may be offset by the creation of additional exports or reexport of the assembled finished goods or by the recorded reinvestment of profits, they may give rise to a net drain on reserves since the FDI flows do not represent actual foreign currency inflows or the foreign receipts may be used to meet claims of foreign investors, while the imports will always give rise to claims on foreign currency reserves.

[Other participants at the round-table noted the positive effects of FDI in the shape of access to foreign technology and foreign exchange earnings related to exports from the FDI. But whether all these have positive overall benefits would depend on the ability of the government to control FDI, and direct the FDI to priority sectors and determine the conditions for entry and establishment of the FDI, including technology transfers.]

But the most important aspect is related to the reinvestment of earnings from FDI investments. Just as any Ponzi-type scheme, it works as long as the reinvestment continues and foreign investors are willing to take their profits in terms of increased domestic investment. The problems occur when the inflows are not sufficient, or when the foreign investor decides to reallocate capital to an alternative investment location.

It is here that the wonder of compound interest plays a crucial role. At an annually compounded rate of return of 10%, investment value doubles in a little over 7 years, at 15% it doubles in under five years, at 25% it doubles in just over 3 years. A country with a tiger-like growth rate of 10% per annum and an initial net FDI inflow of 10% of national income would find itself with an accumulated stock of FDI equal to national income in under 18 years if foreign earnings are 25% (at foreign earnings of 20% the time rises to about 26 years, if the growth rate is 7.5% it falls to 15 years). This is without any FDI except the reinvested earnings on the original stock. Thus, by the end of the period of 18 years, foreign claims with respect to profits would be equal to 25% of GDP. On the other hand, if new FDI remains at 10% of GDP, so that it grows in step with GDP, then the accumulated FDI stock reaches 100% of GDP in around 10 years.

Aside from any other issues, if we assume that the country has settled into an equilibrium on its external accounts which accommodates this FDI inflow, a structural instability is built into the system. This can be seen from two possible scenarios.

In the first, a crisis may be generated without any necessity of capital outflows through the sale and expatriation of capital invested in bricks and mortar. It is sufficient for a domestic or international capital market event, to cause foreign investors to stop their FDI inflows. Just as a modern bank run occurs not because depositors withdraw their funds but because lenders refuse to renew their loans to the bank, a developing country which has adjusted to the permanence of FDI flows will be increasingly exposed to a chance interruptions to those flows which is completely independent of the permanence of the previous flows embodied in the existing stock of FDI.

In the second scenario, assume that foreign investors decide, when the ratio of FDI stock to GDP reaches 100%, that it would be prudent to start to reduce their exposure and start to repatriate 50% of annual profit flows. Again, without any necessity to liquidate current holdings, this would mean additional annual export earnings of 12.5% of GDP would be required to service this profit repatriation, irrespective of imports. It would also mean that the export surplus would have to expand more rapidly than the domestic growth rate if an exchange crisis were to be avoided. This is clearly unsustainable, both in terms of the share and the rate implied. The share and the rate of increase in the export surplus which would be required to support even a 25% profit repatriation ratio would be unsustainable unless the net contribution of FDI investment to exports was equivalent to profit repatriation.

These are the sorts of structural impacts which excessive reliance on FDI flows may produce on the payments flows of the economy and highlights the similarity to a Ponzi scheme of FDI flows based on the high reinvestment of profits. The structural instability that arises is independent of any short-term instability of FDI flows due to the financial innovations mentioned above and would arise even if FDI is as permanent as is traditionally supposed.

Unless FDI flows are truly permanent, in the sense that neither profits or principal are repatriated, the more successful a country is in attracting FDI, and the more successful that FDI is in terms of generating returns, the more fragile will be the countries current account position and thus its exchange rate. But, both these factors will increase the currency risk of the FDI and lead to the increased probability of repatriation or hedging through the foreign exchange market. If success also raises domestic incomes and costs, thus reducing domestic rates of return, this will reduce the size of FDI reinvestment flows and take pressure off the current account, but it will also lead to greater attractions to shift investments to other locations and thus to much more massive shifts in invested capital.

Developing countries may thus find themselves in a position which resembles that of the United States in the 1960s with respect to its gold reserves. European countries had built up large investments in dollar reserves on the understanding that they were convertible into gold. But, the outstanding dollar claims soon exceeded the US gold supply... in the end, (when) even the permanent claims started to be exercised, the result was a collapse of the dollar and the suspension of gold convertibility.

The problem for a developing country is to assess what proportion of FDI flows are indeed permanent, and what their short- and long-term impact on trade flows will be. It is probable that the higher the return on investment and the higher the proportion of reinvested earnings in total FDI stock, the less permanent the FDI stock will be and thus the greater the threat to the BOP and to exchange rate stability. In this respect the choice between FDI and other types of foreign borrowing is one of degree rather than difference, and the amount of foreign investment, whatever its nature, cannot be a matter of indifference to government policy.

Thus while portfolio flows may have a more direct impact on short-term reserve management and exchange rate policy, FDI may have both a short-term and a more long-term, structural influence on the composition of a country's external payments flows. While financial innovation allows FDI to have an impact in the short-term which is increasingly similar in terms of volatility to portfolio flows, the more important aspect is the way it may mask the true position of a country's BOP and the sustainability of any particular combination of policies. Just as accumulated dollar claims on gold represented a Damocles sword hanging over the gold-dollar exchange rate, accumulated foreign claims in the form of reinvested profits may create a potentially disruptive force which can offset any domestic or external policy goals.

During the 1960s and 1970s it was common for countries such as Germany to place direct controls on capital inflows, just as it was common to regulate the raising of capital in domestic markets by the US, Switzerland, Holland and other developed countries. Current conditions do not suggest that developing countries should be prevented from using the same types of regulations to protect their domestic and external stability.

(Prof. Jan Kregel teaches international economics at the University of Bologna, Italy. The above article is based on, and abstracted from, his comments on implications of financial globalization for development policy, at a seminar on East Asian Development in Kuala Lumpur, Malaysia, in March)