Flexible Exchange Rates and the Domestic Savings Rate


In the past few years, a new theory is being developed with respect to the level of the exchange rate under a system of freely floating exchange rates or even under fixed rates.

ByAntonio Carlos Lemgruber

More and more, economists are explaining situations of overvaluation or undervaluation of the exchange rate of a country through a direct link with domestic savings. To use two excellent examples: it is argued that China can sustain an undervalued exchange rate because the country has a huge level of domestic savings; in contrast, the exchange rate would tend to be overvalued in Brazil because of the very low level of domestic savings.

Naturally, this linkage between exchange rates and domestic savings is derived from certain macroeconomic identities: the current account of  the balance-of-payments corresponds to external savings. Consequently, China would have a tendency for surplus in the balance-of-payments current account, as opposed to Brazil which needs to generate external savings and therefore to produce a current account deficit. In simple terms, Brazilian exports of goods and services may – or should – receive less incentives as far as the level of the exchange rate is concerned than, for example, Chinese exports of goods and services. Behind this concept, one should consider the enormous difference with respect to the domestic savings rate in these two countries.

The causation story in this theory runs as follows: in order to reach an adequate level of total investments, countries with low domestic savings need to produce a deficit in the current account of the balance-of-payments and consequently such deficit requires an overvalued exchange rate. And vice-versa for countries with high domestic savings. In other words, trade surpluses and export growth are good for China and bad for Brazil.

Some interesting discussions about the link between exchange rates and savings rates can be found in Mr. Google:  just search for these macroeconomic terms. Specifically, we make reference to the presentation of Ben Bernanke at the ChineseAcademy of Social Sciences ( Beijing, China) in December 15, 2006 – The Chinese Economy: Progress and Challenges.

This is of course a very dangerous new theory, with very serious implications for economic policy decisions. For instance, focusing on our two examples – Brazil and China – the theory provides arguments to indicate that one should not worry neither about the Brazilian overvaluation nor the Chinese undervaluation. Brazil supposedly needs current account deficits and China needs current account surpluses, given their domestic savings rate.

It is not the purpose of this paper to explain why domestic savings are so different in these countries, although it seems to be obvious that this has a lot to do with the existence in Brazil of a major social security system, as opposed to China. On the other hand, it is the purpose of this paper to suggest that such theory certainly has some flaws that must be pointed out, particularly because of the growing importance of this exchange rate-savings rate link in the economic literature.

It is true of course that a devaluation of the Brazilian exchange rate would reduce the current account deficit and eventually turn into a surplus – and vice-versa for China, that is, an appreciation of the Chinese currency would reduce the current account surplus, by affecting exports and imports of goods and services.

What is questionable is the next step of such theory, that is, the argument that a lower current account deficit – corresponding to lower external savings – will necessarily reduce the total investments of a country. And vice-versa for a smaller current account surplus turning into a deficit eventually through appreciation of the currency (the Chinese case).

It is very common, when we deal with identities, to make the following mistake. If A = B+C, and if B falls, then A necessarily falls. This is simply not true. A might stay constant – or even grow – if the decline in B is compensated  by an increase of C. Just replace A for total investments, B for external savings ( current account deficit), and C for domestic savings and such substitution hypothesis will be well understood. In other words, a macroeconomic identity should not be transformed into a theory.

What must be pointed out is the clear possibility of a reverse causation factor. In other words, a devaluation of the exchange rate might very well provoke an increase in domestic savings. One important channel of transmission is the following: an overvalued exchange rate represents a subsidy for nationals to buy foreign exchange and spend abroad (think of tourism or imports) and also for foreigners to avoid purchasing the domestic currency and – worse – to repatriate funds away from the country. A devaluation, after all, eliminates this subsidy.

A competitive exchange rate will avoid consumption spending booms and increase profits in the tradable sector. A competitive exchange rate stimulates investments, exports, profits and thus domestic savings.

In our view, there are flaws in the theory that correlates low domestic savings with overvalued exchange rates. As it happens in many cases in economics, the causation probably goes the other way. Undervalued exchange rates are helping China to build domestic savings designed to finance investments domestically and abroad. Overvalued exchange rates are hurting Brazil, by leading to current account deficits that tend to displace domestic savings and might have serious consequences a few years from now (just like it happened many times in the last century).

This is a subject that certainly deserves further theoretical and empirical discussion, particularly in 2010 and the following years, due to the Chinese currency undervaluation and to many cases in emerging markets – particularly Latin America – of clear and dangerous overvaluation of exchange rates.


Emerging Markets