This Background Paper examines sources of net international reserve inflows for Poland. It shows that once estimate of unrecorded border trade is included, both the current and capital accounts are in surplus. The paper illustrates the methods through which estimates of unrecorded trade can be obtained. It offers a complementary view of Poland’s inflation problem from a medium-term perspective. Specifically, the paper shows that relative price adjustment, combined with indexation and inertia, has been part and parcel of Polish inflation.

Abstract

This Background Paper examines sources of net international reserve inflows for Poland. It shows that once estimate of unrecorded border trade is included, both the current and capital accounts are in surplus. The paper illustrates the methods through which estimates of unrecorded trade can be obtained. It offers a complementary view of Poland’s inflation problem from a medium-term perspective. Specifically, the paper shows that relative price adjustment, combined with indexation and inertia, has been part and parcel of Polish inflation.

III. Policy Responses to the Reserve Inflow Problem

There is a consensus that excessive reserve accumulation is a central part of Poland’s present inflation problem. However, behind this consensus lie different views on the causes of the reserve accumulation, and this has lead to potentially conflicting policy advice.

These different views were articulated in Section II of the Staff Report for the 1995 Article IV Consultation (SM/95/310, 12/14/95). This chapter attempts to reconcile these views, to the extent possible, within a common macroeconomic framework. The chapter concludes that, despite the deep disagreements on the causes of reserve accumulation, the policy implications need not be that dissimilar. To be more precise, regardless of the proximate cause of the reserve accumulation, some combination of exchange rate appreciation and fiscal adjustment emerges as a general policy prescription. Uncertainty over the causes of reserve accumulation, and the likelihood that the relative importance of the underlying causes itself will change over time, strengthen this conclusion.

As described in the staff report, there are (at least) five competing interpretations of Poland’s problem of reserve accumulation and inflation.

A first view is that monetary policy is too loose. However, while loose monetary policy could explain Poland’s inflation problem, it cannot explain the accumulation of international reserves. Rather, the loose monetary policy is the consequence of excessive reserve accumulation. According to this view, inflation reduction depends on lower money supply growth and higher interest rates. Proponents of this view would tend to argue that the simplest way to restore control over the money supply would be to allow the exchange rate to float freely. This would ensure overall payments balance, remove external pressures on the money supply, and allow money supply targets to become the new nominal anchor.

While this is certainly a coherent strategy, this approach is not without practical problems. Barely four years of post-transition, post-hyperinflation data make it hard enough to estimate a money demand equation, let alone be convinced of its stability amid structural changes in the economy. Indeed, the legacy of widespread dollarization and the potential for currency substitution may make control of the appropriate monetary aggregate impossible at this time. Finally, this view does not address the underlying causes of the increase in international reserves. Thus, even if the policy implications are sound, this view does not offer a comprehensive explanation of Poland’s reserve inflow problem.

For these reasons, and to best relate to the policy options entertained by the authorities, the rest of this chapter takes the current, essentially fixed exchange rate regime as given, and examines the policy options within this institutional constraint. To aid the exposition, a simple macro-economic model is used. This helps to clarify the differences between various views, and to ensure internal consistency in each of them.

The standard Mundell-Fleming open economy model is used for the analysis (Chart III.1). For ease of exposition, we assume that Y* represents the economy’s equilibrium level of output supply. Clearly, in a restructuring transition economy such as Poland, Y* should increase over time--so we need to think of Y* as equilibrium output at some point in time. Furthermore, we suppose that there is excess demand if actual demand, Y, exceeds equilibrium output supply, Y*. Finally, we suppose that such excess demand leads to higher inflation. 1/ For the most part, the IS and LM curves have their conventional slopes; for simplicity, the BP curve is either horizontal (perfect capital mobility) or vertical (zero capital mobility). 2/ If the BP curve is horizontal, the domestic nominal interest rate equals the sum of the foreign interest rate and the expected rate of exchange rate depreciation, plus a shift term, reflecting a risk premium demanded on Polish assets. Points above or to the left of the BP curve represent balance of payments surplus, those below or the right a payments deficit.

CHART III.1.
CHART III.1.

POLAND THE EFFECTS OF INCREASED MONEY SUPPLY

Citation: IMF Staff Country Reports 1996, 019; 10.5089/9781451831788.002.A003

With this apparatus in hand, a more detailed discussion of alternative interpretations of Poland’s problem of inflation and reserve accumulation is possible.

In the staff report, the second view mentioned is that monetary policy is too tight. This view can explain the increase in international reserves, but not why inflation remains so high. Nevertheless, advocates of this view argue that lower interest rates will solve the problem of inflation and reserve inflows. By reducing the differential with foreign interest rates, lower interest rates can solve the capital inflow problem. By increasing domestic absorption, lower interest rates reduce the pressure on the domestic money supply from the current account surplus.

In one sense this view is uncontroversial. If lower interest rates were simply the consequence of measures such as fiscal consolidation or nominal appreciation--both of which reduce demand--there would be no increase in inflation. However, this view becomes more controversial if it is interpreted as an argument for a money supply increase.

An increase in money supply lowers interest rates by shifting the LM curve out and to the right. The lower interest rates would reduce capital inflows but, by stimulating demand, would add to domestic inflationary pressure. This would only be appropriate from an initial position of deficient demand (Chart III.1). However, it is difficult to reconcile deficient demand with the existence of inflationary pressure. True, inflation may be inertial, or demand may be deficient in some sectors; even so, a money supply increase and lower interest rates would still add to inflationary pressure. Alternatively, if expenditures are interest-insensitive (vertical IS) money supply expansion would solve the reserve inflow problem without risking inflation, but only by assuming that monetary policy has no effects on aggregate demand. Thus, in most cases, increasing the money supply to lower inflation should only be considered when part of a package of fiscal consolidation or exchange rate appreciation.

The final three views are somewhat less controversial. Indeed, they can be seen as complementary explanations of Poland’s reserve inflow and inflation problem. Concerning the present state of the economy, each of the three explanations agrees that there is excess demand (Y>Y*) and a balance of payments surplus (the economy is to the right or above the BP line); i.e., the present state of the economy is in the north-east quadrant of the IS-LM diagram (point A in Charts II.2 and II.3). Where they differ is on how best to return the economy to macroeconomic balance.

The third view described in the staff report sees the source of Poland’s reserve accumulation in inconsistent macroeconomic policies, specifically, tight monetary and loose fiscal policy. It is argued that, in part, this policy mix reflects an attempt by macroeconomic policy to overcome the effects of inertial inflation and widespread indexation. There is a policy conflict between the high nominal interest rates needed to ensure positive real rates, and the lower nominal interest rates compatible with payments balance.

The situation is represented in Chart III.2. Assume there is full capital mobility, and that the economy is initially at point A. Fiscal contraction takes the economy to point B: interest rates fall to international levels, removing the balance of payments surplus, but demand falls below Y*, the equilibrium level of output supply. To restore full macroeconomic balance to point C, some combination of fiscal contraction and looser monetary policy is required. 1/

CHART III.2
CHART III.2

POLAND THF EFFECTS OF FISCAL CONSOLIDATION

Citation: IMF Staff Country Reports 1996, 019; 10.5089/9781451831788.002.A003

An alternative interpretation of this third view would stress the policy conflict for domestic interest rates between establishing domestic and external balance. Suppose that inertial inflation is such that nominal interest rates have to be held high to preserve positive real rates. Now suppose that the risk premium on Polish assets has fallen because of increased foreign investor confidence. Alternatively, consider a reduction in the rate of crawl, and assume that goods markets and expected inflation are slow to adjust whereas financial markets adjust rapidly. In both cases the BP curve will be well below the interest rate consistent with the inflation rate.

If there is perfect capital mobility, the position of the BP curve ultimately determines the interest rate. The issue is how to choose macroeconomic policy compatible with this lower interest rate, while preserving internal balance. By shifting the IS curve back to the left, fiscal consolidation and nominal appreciation can offset the stimulus to domestic demand from the lower interest rates.

Fiscal consolidation has its own complications. The size of fiscal adjustment needed to lower interest rates sufficiently to stem the capital inflows may be politically unrealistic. Even if carried out, such a policy may prove a victim of its own success: by lowering the relative supply of domestic assets or by signaling the government’s fiscal credentials, the risk premium on domestic assets may fall as a result, shifting the BP curve further down, and necessitating additional fiscal consolidation.

A fourth view sees the root of Poland’s reserve inflow problem in an undervalued real exchange rate. The fundamental causes of this undervaluation may be Poland’s rapid productivity growth, and improvements in the quality of its exported goods, each of which would imply trend appreciation of Poland’s equilibrium real exchange rate. In addition to these dynamic processes, this undervaluation may have been exacerbated by the devaluation of mid-1993.

According to this view, it is the real exchange rate disequilibrium that has compromised the effectiveness of the exchange rate as a nominal anchor. The solution is nominal exchange rate appreciation to remove the disequilibrium and to allow the nominal anchor to bite. In a world of complete goods market integration and flexible prices, such a disequilibrium would be removed instantly by step increases either in prices or in the nominal exchange rate. However, in a world of fixed exchange rates and slow price adjustment (as in the Mundell-Fleming model), disequilibrium is resolved by current account surpluses, reserve inflows, and increases in the money supply. Prices rise until the real exchange rate appreciates to its equilibrium level.

The macroeconomic implications of this view can be illustrated in the Mundell-Fleming framework, with or without capital mobility. As mentioned above, exchange rates are assumed fixed. In Chart III.3, improved competitiveness has boosted net exports, shifting both the IS and BP curves out and to the right. 1/ There is a balance of payments surplus, which increases the money supply. The resulting rightward shift in the LM curve will take the economy to point B, with payments balance restored. 2/ With complete capital mobility, sterilization is impossible and the movement to B is immediate; with imperfect capital mobility, the incipient money supply increase can be restrained through sterilization. Either way, output rises above its equilibrium level, Y*. 3/

CHART III.3
CHART III.3

POLAND THE IMPLICATIONS OF AN UNDERVALUED EXCHANGE RATE

Citation: IMF Staff Country Reports 1996, 019; 10.5089/9781451831788.002.A003

Over time, excess demand raises prices, shifting the IS, LM and BP curves leftward, back to the initial equilibrium. The initial competitiveness improvement is ultimately offset by real exchange rate appreciation, achieved through higher domestic prices.

In such circumstances, the first best policy response is nominal appreciation. The IS and BP curves shift leftward, restoring macroeconomic balance. The required real exchange rate appreciation is achieved through immediate nominal appreciation, not the drawn out process of reserve accumulation and higher prices.

In a world of capital mobility, fiscal contraction would also help restore internal macroeconomic balance. From A, the IS curve would shift left, offsetting the boost to demand from enhanced competitiveness. Interest rates fall, checking the incipient capital flows that--if left alone--would have raised the money supply. However, while internal balance is restored, the resulting import compression adds to net exports. Indeed, in the extreme case of zero capital mobility, fiscal contraction is counterproductive. It would shift the IS curve toward equilibrium, but not the BP curve. By reducing imports, fiscal contraction would only add to the balance of payments surplus and the pressure on the money supply to increase.

In the absence of capital mobility, reducing the rate of crawl comes close to the first-best policy of step appreciation. Competitiveness is slowly worsened--relative to the baseline--over time, pushing the IS and BP curves back to the left and checking inflationary pressure. Indeed, in the limit--a simple monetary model with a single tradable good--domestic inflation is simply foreign inflation plus the rate of crawl. However, in a world of capital mobility where goods market integration is incomplete, lowering the rate of crawl can have perverse effects. As described under the third view, by increasing the prospective rate of return on domestic relative to foreign assets, a lower rate of crawl would encourage capital inflows.

Some additional complications can be incorporated. In the presence of inertial inflation, either a step appreciation or reduction in the rate of crawl mean that the real exchange rate will slowly appreciate over time, perhaps putting competitiveness at risk. In addition, if expectations are extrapolative, step appreciation could increase the attractiveness of domestic assets, shifting the BP curve further down and adding to reserve inflows and the downward pressure on domestic interest rates. Conversely, if exchange rate expectations are Man-reverting (perhaps to some (changing) “equilibrium” real exchange rate), exchange rate appreciation could raise the expected rate of future depreciation. The BP curve would then shift up, allowing the higher domestic interest rates needed to address the problem of domestic inflation. Arguably, the more aggressive the appreciation today, the more likely an upward BP shift, and the greater the freedom for domestic interest rates to increase. However, as discussed in Chapter 2, this freedom comes at the cost of risks to competitiveness.

Finally, a fifth view argues that the source of the increase in reserves lies in Poland’s increasing attractiveness as an emerging market for foreign investment. Proponents of this view point to Poland’s improved access to international capital markets, its recent successful Eurobond issue, and the increase in foreign direct investment.

This is essentially a dynamic view, and it is difficult to incorporate the full effects within the confines of the static Mundell-Fleming model. Nevertheless, in simple macroeconomic terms, the effect is to lower the BP curve over time, forcing domestic interest rates down.

On this view, the question is how best to harness these capital inflows into productive investment, without overheating the economy. At issue here is the nature of the capital flows (medium term or speculative), the amount of slack in the economy, the availability of profitable investment projects (and the capacity of the financial system to transform inflows toward them) and the question as to whether the flows will finance investment and not an unsustainable consumption boom. From these considerations follows the appropriate policy stance: how much sterilization, how much appreciation, how much fiscal consolidation, and the question of capital controls. 1/

To conclude, for Poland, it has been difficult to determine with precision the relative magnitudes of the current and capital account surpluses. 2/ Indeed, in practice, the two may be closely linked in so far as improved current account performance is likely to increase investor confidence and the international demand for domestic assets. However, when dealing with the Inflationary effects of reserve inflows, the real issue is not so much whether the problem is capital or current account, since both fiscal adjustment and exchange rate appreciation can lower inflationary pressure and restore internal balance. Also, by allowing lower interest rates, both fiscal adjustment and exchange rate appreciation can offset the pressure toward capital account surplus. Given uncertainty over the causes of the reserve increases, and the prospect that these causes will themselves change over time, some combination of fiscal consolidation and exchange rate appreciation will be at least part of the optimal policy mix to reduce inflation.

If inflation reduction is the priority, the difficult question is how best to choose between fiscal and exchange rate policy. The answer will depend less on the absolute sizes of the present current and capital account surpluses, and more on the balance between the traded, nontraded and government sectors. Questions here include the present state of health of these sectors, as well as what their relative sizes would, ideally, be in the future. This requires using the right mix of fiscal and exchange rate policies to attain internal and external balance.

1/

Even these assumptions, standard for industrialized economies, may be problematic in the case of Poland. First, aggregate output may not be the most useful concept. As Poland restructures, some sectors may be expanding and others contracting. Excess demand in newly emerging sectors may be accompanied by deficient demand in declining sectors. Thus, there could be excess demand in aggregate, yet considerable excess supply in certain sectors. Also, the equilibrium in the economy will change over time. Second, widespread wage and price indexation mean that inflation has a considerable inertial element. True, “excess demand” will still lead to inflation, but only over and above what is already in the pipeline. The aim of this chapter is to present a simple framework for understanding the macroeconomic issues, upon which the implications of these profound structural issues (in part discussed in Chapter 2) can be superimposed.

2/

The implications of the effects of moderate capital mobility can be inferred from these two cases.

1/

Note that, with zero capital mobility, it is difficult to do justice to this view. The reason is that any excess domestic demand and inflationary pressure stemming from loose fiscal policy would cause balance of payments weakness. and not reserve inflows. However, the third view takes on increasing relevance as the degree of capital mobility increases.

1/

When the BP curve is horizontal (perfect capital mobility), rightward shifts leave its position in the chart unchanged.

2/

With zero capital mobility, increased imports exactly offset the initial boost to net exports; the initial improvement in competitiveness ultimately has no effect on the trade balance. With perfect capital mobility the increase in imports only partially offsets the initial increase in net exports.

3/

This treats the competitiveness improvement as equivalent to a real depreciation. If it was the result of productivity gains, the equilibrium output level, Y*, would also increase though presumably, because of multiplier effects, by less than the rightward shifts in the IS and BP curves. To capture these effects, a more dynamic--but more complicated--model is needed.

1/

These issues are considered in detail in Susan Schadler et al., “Recent Experiences with Capital Inflows”, Occasional Paper No. 108, (Washington: International Monetary Fund, 1993).

2/

See Box 2 of the staff report and Chapter 1 of this document.

Republic of Poland: Background Paper
Author: International Monetary Fund