Real exchange rates have soared during the transition from central planning to market economies. New research by Charles Wyplosz and his colleagues helps us understand these dramatic increases.

The appropriate exchange rate regime for formerly command economies remains an issue of controversy, following the recent attack on the Czech currency, speculative pressure on Poland and Russia, and the adoption of a currency board in Bulgaria. Yet little is known about the specific forces driving the exchange rate during transition.

In 1995, László Halpern and Charles Wyplosz proposed a series of ‘stylized facts’ concerning these exchange rates, notably that on returning to convertibility, the currencies in transition economies underwent a massive devaluation followed by a path of real appreciation, sometimes by impressive amounts. These ‘facts’ have been confirmed by later research but to date, there have been few efforts to explore their theoretical underpinnings.

A new CEPR Discussion Paper (also by Charles Wyplosz, this time with Clemens Grafe) offers a first step in the explanation. It models the transition process as one where old state-owned enterprises are replaced by a modern industrial sector and a previously undersized non-traded (service) sector. The analysis recognises the crucial role of external finance while the domestic financial system is being built from scratch. It also takes into account limited labour mobility out of the subsidised old state sector.

The central result of the paper is that the real appreciation of the exchange rate plays a crucial role in bringing about the reallocation of labour away from the old state sector. Real appreciation allows the new sectors to attract manpower by raising the real wages they offer.

This result contrasts with the so-called Balassa-Samuelson effect, which predicts that fast productivity gains in the traded goods sector are accompanied by real appreciation. In the Grafe-Wyplosz model, causality runs the other way: real appreciation and increases in real wages force the old sector to raise its productivity, which effectively leads to the closing down of inefficient units. Such an effect explains why ‘old industrialists’ complain about overvaluation - and why they should be ignored.

In addition, the paper looks at state subsidies and foreign financing, both of which may have a role in the reallocation of labour, though they may also lead to ‘transition traps’: situations when the real exchange rate is either too low to force a scaling down of the old industrial sector or too high to allow modern firms to emerge. Just as undervaluation blocks transition by preventing labour from migrating to the modern sectors (traded and non-traded goods), overvaluation may wipe out profitability in the new manufacturing sector.

The Grafe-Wyplosz model of the exchange rate emphasises the links between productivity, capital accumulation, real wages and relative prices as the old state sector gradually makes room for the modern sectors to expand. The transition process is represented by five main features:

A weak banking system and an unclear transformation of property rights result in the almost complete absence of lending to the new private sector.

Parts of the old manufacturing sector continue to operate under ‘soft budget constraints’. This has the effect of trapping resources in low productivity production lines, many of which are effectively insolvent. The resultinginefficiency affects both demand and supply.

Labour mobility away from the old state sector is limited by various factors inherited from the command economy, including access to housing, health and education. This effect is captured by assuming the existence of a gap between wages in the modern sectors and wages in the old state sector.

Transition is described as the instantaneous release of pent-up demand for services and for internationally traded goods. The result is the emergence of a market-determined real exchange rate - the ratio of the price of non-traded goods to the price of traded goods.

The dismantling of ‘Berlin walls’ is accompanied by the availability of foreign financing.

A number of results emerge from the model. First, the real exchange rate is simply the other side of the coin of the real wage. Labour costs and the real exchange rate need to be initially low to allow the new traded sector to generate high enough profit margins to be able to expand. At the same time, a continuous real appreciation is needed to attract labour away from the state sector, which is then forced to close down inefficient production lines. Contrary to Balassa-Samuelson, rising productivity in the traded good sector is a reaction to the real appreciation, not the driving force behind it.

Second, the proper level of the real exchange rate is a knife-edge. If real wages are too low, they provide no incentives for labour to leave the state sector. But if real wages are too high, they reduce retained earnings and capital accumulation in the modern manufacturing sector. In the latter case, a transition trap occurs: only the old manufacturing sector and the low capital-intensive service sector exist.

Third, frictions in the labour market and subsidies to the state sector enhance this transition trap effect by requiring higher real wages and a higher real exchange rate. Such frictions can be so high that a modern manufacturing sector cannot emerge at all.

Fourth, foreign finance tends to offset the effects of subsidies and labour market frictions. Under certain conditions, it sustains demand and raises the real exchange rate, which, in turn, imposes tougher foreign competition on the old sector. Alternatively though, limits on foreign borrowing can cause the real exchange rate to be undervalued (given the productivity levels), thus keeping real wages too low.

Finally, if there is too much foreign financing, it can lead to another transition trap, one akin to the Dutch disease: a large supply of foreign funds props up demand, which leads to real appreciation and high real wages. This wipes out profits in the modern traded sector and, given the financial market distortion, blocks investment and the development of state-of-the-art manufacturing.

The Grafe-Wyplosz results have a number of policy implications. One is that it is futile and possibly counter-productive to resist real appreciation since it is the market channel through which labour is attracted out of the inefficient state sector. Another is the crucial role played by IMF or World Bank money in the early days of reform: in the absence of other outside sources, such financing allows countries to overcome the debilitating effects of distortions in the labour and financial markets.

This is true even though the model assumes that foreign loans are used purely to finance consumption. Investment rather than consumption of foreign funds is naturally preferable since they supplement retained earnings in the build-up of capital stock and hence accelerate the transition. But whether used for consumption or investment, capital inflows raise productivity and help the economy outgrow its external debt. This result disproves the popular argument that a consumption boom financed by a current account deficit is harmful.