The notion that one can make inferences about the characteristics of financial flows by just observing their label is not new in economic. There is much convention wisdom that show capital flows reflect speculative, unstable behavior while flows reflect evaluations of long run profitability and are based on fundamental economic condition. The flows of funds approach used by many central banks and others for a analysis of the domestic economy developments is based on labels which are deemed meaningful.
This view has also been an important part of the traditional analysis of international finance for many years. In fact, the structure of balance of payments accounts reflects an implicit theory that different types of capital flows have different economic implications. For example, the distinction between short-term “hot money” and long-term capital flows undoubtedly reflects the view that short-term capital movements are speculative and reversible while long-term capital flows are based on fundamentals and are reversed only when the fundamentals change. The fact that capital control programmes in many countries distinguish between short-and long-term positions also points to the importance attached to this distinction. Another important distinction found in balance of payments accounts is between official capital flows, including changes in international reserves, and private capital flows. A common view in the context of current capital inflows into developing countries which is based on labels is that such flows are fundamentally different form inflows to these same countries in the 1970s because the current inflows are private-to-private transactions not guaranteed by the government of any country.
Finally, it is often argued that direct investment has different implications for the host and recipient countries as compared to other capital flows. For example, direct investment capital flows are often associated with technology transfers and a range of costs and benefits for both countries. An implicit assumption behind these ideas is that the transactions reported in the balance accounts are closely related to the behaviour of interest in the real economy. This reasoning based on the label of the flow at times underlies substantive policy measures. Once a flow is identified as “hot money,” it is then seen as requiring some policy response. At various times countries (especially developing countries) have responded with exchange rate management, (sterilised) intervention, fiscal contraction, borrowing taxes, absolute foreign borrowing constraints, and reserve requirements. There are a number of good reasons to doubt, however, that the micro nature of international capital flows reveals much about the economic importance of such flows. Since many assets are (increasingly becoming) tradable, a distinction between flows based on their terms (for example, short versus long) is (increasingly becoming) less meaningful. A treasury bond with a 30-year maturity can easily be more liquid, and thus lead to a higher volatility of short-term flows than a 30-day time deposit at a commercial bank. And, a short-term asset, which is, rolled over can in many ways he identical to a long-term asset.
Furthermore, the explicit label given to a flow may not cover its implicit nature. For instance the inflows to developing countries in the 1970s were private capital flows in name, the universal government guarantees of both lenders and borrowers considerably subdued the discipline of the market. In effect the capital flows that helped generate the debt crisis of 1982 should have been considered official capital flows since the private parties undertaking the transactions relied on a government guarantee (and ex post, private claims indeed became the liability of the government). This presumably made private investors less careful than they would have been in the absence of guarantees. This experience should act as a warning against evaluating capital flows according to their label (for example, the instrument traded or the transactor recorded in the balance of payments statistics). In the past few years inflows to developing countries have taken the form of non-guaranteed portfolio investments and direct investments. If the behavior behind such investments is different in some important way, it follows that we may not be inviting another debt crisis even if conditions change as they did in 1982. But if the flows are to a significant extent guaranteed by the government, their label “private” may be meaningless. Clearly, in thinking about stability, an important attribute is thus the contingent liability of the government.
The starting point should then also be a clear methodology on how the hypothetical economic implications of different types of capital flows can be identified empirically. If credit markets were perfect and complete, the form of capital flows would not be important. A useful analog here is the Miller-Modigliani theorem from corporate finance. Under a set of strong assumptions, the structure of assets and liabilities among various types of debt and equity have no effect on the value of the firm because investors can offset the structure chosen by the firm in credit markets. If we think of the balance of payments accounts as records of how a country finances its international capital position it follows that under the Miller-Modigliani assumptions, the structure of capital flows and the structure of the gross international asset and liability positions would be unrelated to the country’s net indebtedness.
It might be useful here to look at an analogous issue that has recently been carefully explored in a closed economy context. That issue is whether a “credit crunch” contributed to a downturn in economic activity. This is an interesting hypothesis because it explicitly rests on the view that one type of financial transaction, in this case loans by domestic banks to domestic non-financial firms, had important economic effects.
This view is an important part of the new literature on business cycles, which links financial structures and real activity. This literature provides a useful analytical framework that can be adapted to a discussion of international financial flows. It points out that the importance of changes in bank credit for economic activity, as opposed to bank liabilities or money, is an empirical issue. If good substitutes for bank credit exist in an economy, it follows that a decline in bank credit that is not matched by a decline in money will have no effect on economic activity. Firms would easily substitute other forms of credit, for example commercial paper, to offset the reduction in bank credit. In contrast, if banks have special information about their customers that is not easily transferred to other lenders, a reduction of bank credit will not be easily offset by borrowing from other institutions or markets. In this case an interruption in bank credit could have a depressing effect on expenditures and output.
There are two important lessons from this. First, meaningful tests of the importance of capital account transactions require a specification of the economic behavior of interest. Second, a financial transaction is likely to have measurable effects on the specified economic behavior in cases where the institutional environment, information structures, or other departures from complete markets limit the ability of investors to substitute one type of transaction for another in response to changing incentives.
The view that labels matter can take comfort from the fact that international financial markets are not complete and transactions are subject to a large number of distortions. Many of these are imposed by governments in the form of controls on types of transactions that are considered undesirable. Moreover, subsidies often take the form of a government guarantee of private liabilities, favorable tax treatment on earnings (direct investment), or access to special government facilities (debt equity swaps). Each of these distortions is designed to encourage or discourage a given type of capital transaction. Furthermore, governments also intervene directly in international capital markets. Developing country governments borrow for long-term objectives and, increasingly, in middle-income countries, to offset short-run pressures on exchange rates and domestic interest rates. A difficulty thus arises in interpreting private capital flows that are sterilized by intervention transactions by the central bank as, to some extent, the composition of private capital flows is conditioned by governments’ capital transactions. Clearly a country trying to maintain interest rates above its trading partners in some sense generates the private capital inflow by standing ready to match private inflows with official outflows in the form of increases in reserve assets. Such private inflows are “sustainable” as long as the official capital outflows are “sustainable.”
In general, if international capital transactions are effectively distorted by taxes, subsidies and direct intervention, the structure of private capital flows might follow predictable patterns as long as the structure of distortions itself is stable. Over a period the financial transactions are motivated by a variety of economic forces. Over time the institutional framework changes, the behavior of the official sector probably changes, the exchange rate regime changes, capital control programmes come and go, and banking markets (both on and offshore) develop, fail, and are recapitalized The final economic behavior that seems to provide the most appropriate test for the importance of various capital flows is the net inter trade among countries, conventionally measured by the current account. While there are other interesting candidates, for example, the exchange rate, questions about the sustainability of certain type of capital flows are motivated by concern about the sustainability of a path for changes in net indebtedness to the rest of the world, that is, the mirror image of the cumulative current account balance. Policymakers wish to assess the likelihood of sudden and destabilizing changes in the total capital account not just its components.
On a macro level, national accounts identities state that the sum of all capital flows is equal to the difference between savings and investment. For a given savings investment imbalance, capital flows have to satisfy an adding-up constraint. Unless flows influence domestic investment or domestic savings (or a combination of the two), some substitution between the various flows has to occur, for a given current account, the volatility of one flow will on aggregate be canceled out. The extent of interaction between the various components and the possibility of systematic interactions between components thus needs to be addressed before making inferences from the parts to the whole. A fundamental difficulty is that the linkage between current account and any of the capital flows data series might be very weak. For one thing we know that gross capital inflows and outflows are several times larger than net capital flows. Yet it is the net flow that we are interested in when considering sustainability of current account position. It seems clear that the motivation for two-way flows of international capital are very different from the general view that developing countries should be net capital importers and that each type of transaction should show a net inflow.
For example, residents of a small open economy might hold most of their financial wealth in the form of foreign clans since this allows them to diversify their income streams and protect themselves from domestic taxation and income shocks. Offsetting this preference would be the tendency for foreign investors to purchase claims on the country again in order to diversify their risk. This diversification motive for international capital movements might account for much of the recorded flows in balance of payments data. Moreover, it might be the case that residents of wealthy countries prefer long-term investments such as bonds or loans while residents of developing countries prefer short-term investments such as bank deposits. In this case, we would expect to see short-term capital outflow from developing countries matched by long-term loans to residents of developing countries. These preferences imply nothing about the desired imbalance in the capital or current account. In fact, the short-term positions of residents might be quite stable if there are relatively few alternative investments. In contrast, the long-term capital inflow to the developing country might take several alternative forms. In one year foreign investors might prefer equities, in another bank loans, and in another bonds, These preferences might in tutu reflect subtle changes in tax rules (in the industrial countries) that the analyst would find difficult to identify. The point is that the pattern of financial intermediation might be quite stable for a while but that small differences in the institutional framework, due to either regulation or innovation in credit markets, might alter the form that the intermediation taken in balance of payments statistics.
In general, direct investment is a difficult capital flow to interpret. Balance of payments data on direct investment include short-and long-term capital transactions of a loosely defined set of reporters who own more than a given percent of the equity in a domestic or foreign chartered firm. Because direct investors hold factories and other assets that are impossible to move, it is sometimes assumed that a direct investment inflow is more stable than other forms of capital flows. This need not be the case. While a direct investor usually has some immovable assets, there is no reason in principle why these cannot be fully offset by domestic liabilities. Clearly, a direct investor can borrow in order to export capital, and thereby generate rapid capital outflows. In most developing countries, there are Laws against this.
This, so it would be surprising to see negative flows for foreign domestic investment in developing country. There could, however, be offsetting movements in other balance payments accounts or in errors and omissions. On a micro level, there is also much (anecdotal) evidence that, from the point of view of foreign direct investors, flows with very different classifications are actually close substitutes. Multinationals often substitute between and among the various forms of intercompany transfers (retained earnings, provision of and repatriation of capital, and inter-company loans) and loans from local or foreign banks for example, to achieve higher profits net of overall taxes. In other cases, flows may be complementary, that is foreign investment may take place through a combination of FDI and portfolio equity investment.