Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Inflation Targeting in Albania Mimi KODHELI [email protected] “Pushing inflation too low-say, below to 5 percent- may entail a loss of output and seigniorage revenue, suggesting a need for caution in setting very low inflation targets in low-income countries. These countries tend to be subject to larger output volatility and more pronounced price shocks, and program design should take these attributes properly into account. In particular, inflation targets should be set so as to help avoid risks of an unintended concretionary policy stance.” IMF (2006), pg 17 Economics Department - University of Verona Polo Didattico Giorgio Zanotto Viale dell'Università, 4 37129 - Verona - Italy 1. Albanian macroeconomic environment, features and indicators Albania has made substation progress since the beginning of the transition in 1991. Unlike some former Soviet Union and CEE countries that started the liberalization of the economy in the ‘80s (like Hungary e.g.), Albania did not undertake any pre-transition reform before 1991. During 1991-1992 the economy experienced sharp decline in output and high levels of unemployment and inflation. The economic situation led to a mass migration of the world abroad, especially to Italy and Greece. Recovering from the pyramid schemes crises of 1997, Albanian economy continued to experience high positive growth rates. Between 1998 and 2006 real GDP has averaged almost 6 percent annually (Fig.1). Inflation record has also been impressive, varying not far from 3 percent target of the central bank in more than 8 years now (Fig. 2). The vigorous growth led to an estimated GDP/capita of almost US 3,000 in 2006 upgrading Albania to the group of middle income countries (Fig.3). Fig.1 Real GDP growth 1996-’06 (in %)- Source: INSTAT (Albanian Institute of Statistic) Inflation rate 2000 - 2007 in % 4 3.5 3 2.5 3.3 3.5 2 1.5 1 2.5 3.01 2006 2007 2 2.2 1.7 0.5 0 2001 2002 2003 2004 2005 Year Fig. 2 Inflation-annual average 2001-‘07 (in %)-Source: INSTAT Fig.3 GDP/capita in USD-Source: INSTAT The initial rapid speed of growth was mainly observed in those sectors of the economy where market liberalization proceeded quickly like services and agriculture. Despite the initial revitalization of agriculture, lack of investments, land fragmentation and absence of proper land market, land property problems, poor infrastructure reduction market access, and weak food processing capacity, led to a slowdown in agriculture growth to around 3 percent per annum. As a result its share to GDP narrowed to 21 percent from its high 35 percent in 1996 (Fig.4). However, it still remains the main source of income for nearly 40 percent of the population. Fig.4 Contribution to GDP by sectors 1996-2005 -Source: INSTAT In the last ten years sectors like construction and services have been the major contributors to Albanian growth. Services sector is one of the most dynamic sectors of the Albanian economy, constantly providing a high contribution to the economic activity and growth of the country. The weight of this sector to the economy has been increasing, recording a trend expected to continue even in the future. The assessment of performance of this sector is based on the sales index of economic enterprises by services items in wholesale and retail market, hotel and restaurant services, telecommunication and other services. Revenues generated from sales of these services have increased constantly during latest years and currently constitute about 60 percent of overall revenues of economic enterprises. Construction emerged as key sector in sustaining growth, compensating for the agriculture and industry oscillation. During latest years construction has turned out to be one of the main pillars of the Albanian economy, because of its weight in the overall production and the high growth rates evidenced in this sector. Since some years earlier, the annual growth of this sector has fluctuated around 10-15 percent, making this sector be the first in the list of economy sectors for development. Nonetheless, direct and indirect indexes of this sector’s output indicate that further progress requires a broader attention, particularly to a better allocation of financial, private or public resources. Construction developments in recent years, particularly their extreme concentration on certain territories, prove the lack of efficiency of urban development vision, not guaranteeing long-term sustainability of the branch. Furthermore, construction is still dominated by housing building, lagging behind all other strategic items, such as investments in road infrastructure, and other investments directly related to the standard of living of small urban and rural communities. Box 1. Opinions on housing prices in Tirana Nowadays there is accepted the negative impact, which has a bubble effect, of real estate prices on the economic growth. The rapid price rise of houses, not justified by basic economic factors, may lead to inappropriate investments, which will reduce the economy efficiency. The house price, likewise the price of any other asset, influences significantly on the production and inflation in the economy, because when the house price go up, the expectations on their future performance are positive. Such a fact will encourage the increased demand. Also, a higher house price increases the wealth of the individuals, encouraging consumption rise and in consequence, even the aggregate demand. Recently, in Albania, more and more discussions are taking place on rapid price rise of apartments. To study the developments in real estate market and particularly, taking into account the lack of official statistics on this performance, in mid 2006 the BoA carried out a survey on price performance in Tirana house market. 16 real estate agencies, 12 construction firms and 17 commercial banks participated in this survey. The following came out: In the real estate market, the main factor having caused the price rise on the supply side is the reduction of building areas and the land price are of an aboveaverage, having influenced the price rise on the supply side. Essential factors that encourage the demand for living apartments are the increasing demand for house, driven also by demographic movements, the increased preference to invest in long-term periods and increased household incomes (particularly those of emigrants). Most of current transactions for house purchase are based on individual savings. Also the emigrants’ remittances contribute on average by 30 percent to financing the house purchases. Most of sale transactions are carried out in euro, above 90 percent of the cases. Even when payment is made in Lek, the value of house is quoted in euro. Concerning the way the payment is done, the agencies underline that paying by installments represent about 53 percent of the cases, while the rest is made in the form of a down payment. Now days, a considerable part of real estate purchases is financed by the banking system. Most of market stakeholders are of the opinion that bank loans increased in last 2-3 years. Real estate loans extended by the banking system usually constitute 80 percent of the collateral value. All the banks of the system use the market value approach for evaluating the collateral. Also banks apply fixed and variable interest rates for mortgage loans. Together services and contraction count for more than 2/3 of the GDP. Strong domestic demand supported by total factor productivity improvements and remittances has sustained high growth rates during the whole transition period. According to Country Economic Memorandum (WB, 2005) growth accounting analysis in the period 1993-2003 TFP1) growth was mainly the result of resource reallocation from low productivity sectors (agriculture, industry) to high productivity ones (services, construction). However, according to the same report Albania needs to identify other sources of growth to sustain the high rates of growth experienced thus far, such as enhancing capital accumulation and increasing exports. *** The fiscal posture has improved continuously. The size of budget deficit (including grants) was reduced from 10 percent to GDP before 1999 to 3.2 percent in 2006. An ongoing reform to reduce the size of informal economy, which is estimated to be one of the biggest in the region, is expected to increase tax revenues further. The efforts have also reduced debt-to-GDP ratios to just below 55 percent of GDP (Fig.5). But the debt, which is largely domestic, has a short maturity profile and a narrow investor hold base, creating rollover risks. ______ 1) Total Factor Productivity (TPF) growth is measured as a residual, total output growth less the weighted sum of input growth, known as “Solon residual”. Fig.5 Public Debt as a % of GDP Source: BoA Albania has been characterized by large current account deficit throughout the transition period (Fig. 6). Imports have steadily grown reaching 47 percent of GDP in 2007 with exports covering behind at 23 percent of GDP, leaving Albania with a trade deficit of 24 percent. Albania’s receipts from migrant remittances are, relative to GDP, among the highest in the world. Remittances are considerably larger than exports (1.2 times – Sep.07) and represent 14 percent of GDP. Although the volumes of remittances are still sent through informal channels, the importance of the formal channel has increased in recent years (Fig.7). Fig.6 Current Account Balance, 1996-’06 (% of GDP) Source: BoA Fig.7 Distribution of Remittances 1994-’04 Source: IOM (International Organization of Migration) Foreign Direct Investments (FDI) remain low at 3-4 percent of GDP to fill the remaining current account gap which so far has been largely covered with grant and soft loans. While, long run sustainability issues of current accounts may occur, if certain measures that make better the competitiveness of Albanian exports and attract more FDI, are not undertaken. 2. Reasons that pushed Albania to Inflation targeting After bringing inflation down from 237 percent in 1992 to just 6 percent in 1996, inflation has fluctuated around 3-4 percent (with the exception of the limited period after the collapse of pyramid schemes). McNeilly, et al. (1998) attributes the success of Albania in achieving low inflation to the early price liberalization …and an early aggregate supply increase in goods and services alongside a restrictive monetary policy implemented by the Bank of Albania (BoA). As far as monetary aggregates do not contain all the necessary information to predict future inflation, relying on a monetary targeting regime could be suboptimal. (?) Albania has been one of the few countries to have endorsed a monetary targeting regime. Such a regime was appropriate for the framework of Financial Programming upon which the IMF technical assistance programs were based. The monetary targeting aimed to control the money offer in the economy by monitoring M3 as an intermediate target. In compliance with the formal framework of programs supported by the IMF, the monetary targeting was accompanied by quantitative objectives (minimum) of the net international reserve, of net internal means of BoA (maximum) and of the total funding of the budget deficit (ceiling). Such limitations, in addition to being part of the limitations of the IMF programs became operational objectives of BoA through which it would be aimed to achieve control of the monetary supply. Initially, the central bank relied on direct instruments in conducting monetary policy. The supply of broad money (M3) was controlled by ceilings imposed on the level of credit extended by commercial banks. Later on in 1995, the interest rates of State Owned Banks’ deposits (SOBs) were also managed in a first move to switch to indirect instruments. In 1999 credit ceilings were removed, and from 2000 the central bank (BoA ) tries to influence interest rates only through open market operations in T-bills (Repos). Although the regime has been rather successful in terms of price stability, it would not be entirely correct to attribute the success only to this regime. According to Estella and Mishkin (1997), as inflation is brought under control the informative role of monetary aggregates diminishes because the velocity shocks’ relative noise increases. This, according to them, could be an important reason why industrial countries and many emerging market economies do not rely on monetary targeting in the era of price stability. The velocity shocks could be attributed to the instability of money demand which increases as structural changes of the financial system advance. Albania does not make an exception from the rule. Table 1) shows the targeted and actual growth rates of the broad money (M3) and inflation. It is evident that the actual growth rates of M3 until 2000 have diverged widely from the targets. On the contrary, inflation objective during the same period has been most of the time undershot. Even after the year 2000 when M3 growth rates have been fairly close to the targets, they have varied widely, from 5 to 13 percent, compared to the relatively stable inflation rates. This raises doubts about the relative significance of the nominal anchor used in achieving low levels of inflation and makes it very difficult to clearly transmit the intentions of the central bank by simply announcing M3 growth targets. Furthermore, the velocity of M3 has demonstrated a relatively high fluctuation that makes M3 an unpractical indicator in the implementation of the monetary policy. Although the annual growth of the M3 may retain the forecasting capacity, in long term periods of time it is not sufficient to serve on due time to the decision – making relevant to the monetary policy. Consequently, BoA has to look at other possibilities and alternatives of using the information that may serve to implement on due time the monetary policy. The central bank has been aware of this divergence and in practice decisions on changing the monetary policy have been taken based on a wider range of information besides the M3 growth, such as: inflationary pressures coming from exchange rates movements, price changes in different markets, including foreign markets, supply shocks, etc. The experience of transition countries does not provide a clear answer on the choice of an optimal regime of the monetary policy. The best possible conclusion is that the choice should be based on a careful assessment of the specific conditions of the economy and that this choice should be combined with other macroeconomic policies. Sustainability of the overall price level in the recent years gives to BoA the alternative to follow a monetary policy in the form of monetary targeting. While the implementation of the monetary policy is oriented towards targeting of the liquidity level in the market and towards the policy of interest rates in function of keeping inflation low, BoA has maintained flexibility of the response on time towards internal macroeconomic shocks, an example of which was the massive withdrawal of deposits in 2002 (see annual report of BoA, 2002) This way of running monetary policy to some level resembles that of inflation targeting. BoA makes public at the end of each year level of inflation aimed, which led Stone (2003) to classify Albania as an inflation targeting lite country as compared to a full fledged IT country (see Table 2). However, there are two important elements that distinguished Albania from trained and experienced Inflation Targeting regimes: These are the communication of monetary policy strategy including the publication and explanation of its inflation forecast and a formal mechanism that makes BoA accountable to the announced inflation target. Table 2 Inflation targeting Lite Box. 2- Alternative regimes The other alternatives to inflation and monetary targeting are exchange rate pegging and the so called “just do it” policy applied by FED (Mishkin, 2002). Exchange rate pass-through in Albania is thought to be high, though asymmetric, given the large proportion of imported goods in consumer basket (approx. 70 percent). This would make this regime an effective way of conducting monetary policy. The relation of prices to the exchange rate was studied with interest during ’60’70-s. Initially models were established that based on the theory “Law of one price” presumed the existence of a unitarian connection between them. Later on there were developed models of the Parity of Buying Power (absolute and relative) that were used to identify variations in exchange rates. However, numerous empirical tests did not support these suppositions, bringing about the idea that fluctuations in exchange rate are not completely reflected in the level of prices of a country (incomplete pass-through of the exchange rate). Now there are many models/theories that explain why this conduction/conveying) is incomplete. Recently, many empirical studies have testified that in addition to being incomplete, this transmission is also in decline. In literature we meet “first level” pass–through that refers to the sensitivity of import prices in a country towards the variations in the exchange rate of the currency of that country and “second level” pass–through that refers to the sensitivity of consumer prices towards variations in the import prices Nevertheless, many studies do not distinguish between these two stages, including the definition of the effect of the exchange rate fluctuations both the effect in the import prices and in the CPI. Literature on monetary policy in Albania has often considered the channel of exchange rate as the most important one for explaining inflation developments in Albania (Muço et al, 2004). Considering the importance of the exchange rate in a small and open economy as is the Albanian economy, the central bank did neglect the developments in the exchange rate of lek. Such literature has observed that the change in the relation between the exchange rate and inflation after 2000 (Peeters, 2005) where pass–through of the exchange rate with the consume prices in Albania are shown to be in decline. Most of exchange rate discussions, however, refer to scenarios of controlling depreciation to curb higher inflation. From some time, Albania like several other transition countries has experienced appreciation of its currency instead. This has helped easing inflationary pressures. This has pushed BoA to occasionally intervene in the market and to reduce the local currency interest rates against the euro. However, it remains unclear whether this appreciation is a return to some kind of long term equilibrium supported by fundamental changes. This shows that exchange rate developments cannot be ignored that easily in monetary policy decision even if committed to the free floating regime. 3. The transition experience to inflation targeting. Despite of the stable pace of main indicators, the macroeconomic stability is not consolidated yet and it may not be considered that the success of monetary policy has been achieved. The pace of some of the financial market indicators proves that there is still much to do in order to achieve a strong foundation of the Albanian public trust in the national currency. Bolle and Meyers (2005) argue that the inflation prime shown in the yield curve of securities and the high level of dollarisation show that inflationary expectations have not reached yet the level representing BoA objectives. Masson (1997) and Stone (2003) have pointed out the importance of fiscal discipline of the success of the inflation targeting regime. Although the fiscal discipline has been a crucial factor in bringing inflation down in Albania, several events, especially certain structural reforms, have forced the government to slow down its discipline. Among these events an important one is that the government is covering most of the compensation expenses through its budget, and this really put a lot of pressures on inflation. Box. 3 Electricity sector One of the main budget costs for compensation in 2007 is those concerning electricity that amount to 17 -20 million Euros. Unpaid electricity use to be and still is another core problem. The declined pointed in January 2006, when payment was received on only a third of electricity supplied, compared to some 75 percent in the region (IMF country report 2008). Electric power is the main source of energy in the country used by industrial, services and household consumers. Lack of productive capacities, high technical losses, unfavorable weather conditions, and irregularities in bill payments have generated a number of problems accumulated over years. Restrictions in electricity production against the increasing demand have led to re-occurrence of energy crises during 2006-2007, implementing the economic activity in many enterprises. Total electricity production in 2007 declined considerably by 46.9 per cent in comparison to 2006. The total produced quantity was represented 50.1 per cent of the total electricity sources. Electricity import experienced a substantial growth of 136.2 per cent compared to 2006 and represented 49.9 per cent of the total sources in 2007, whereas in 2006 it had represented 18.3 per cent. In 2007 the electricity price in markets grew considerably thus causing the electricity import price to grow over 50 per cent. (Fig. 8) At the same time, power imports jumped three fold to balance the drop in domestic production and poor water reserve management in 2006. Altogether electricity import costs increased by 2 percent of GDP in 2007. Fig. 8 Source: IMF Transition countries have faced difficult alternatives in drafting stabilizing policies. Drop of internal production in the first years following the changes in the political system brought about decline of fiscal income and a rise in the unemployment level and the latter one was also accompanied by increased costs in the form of economic support. Lack of the development of a financial market made the funding of budget deficit the only funding alternative. Amidst natural changes of relative prices during the transition period in a market economy, high deficits contributed directly in high inflation rates. At the same time, the first years were accompanied by a tangible deterioration of the current account. In such circumstance, the economic policy faced parallel challenges of the need to take strict measures aiming to achieve macroeconomic stabilization and of the need to encourage the economy, aiming to reduce unemployment and social tensions. Solving this dilemma was the essence of the pace and dynamics of deflationist policies. The coordination between monetary and fiscal policy in Albania during the transition period has been satisfactory with the international factor playing a critical supportive role in the regard. Since 1992 Albania has gone under several IMF supported programs. The Memorandum of Economic and Financial Policies (MEFP) signed in the course of these programs has been an effective machine to seer both fiscal and monetary policies in a consistent way with the objective of a sustainable growth at low rates of inflation. Modeling and Forecasting Achieving a reasonable accuracy of the inflation forecast, which serves as an intermediate target, is crucial for the successful implementation of the inflation targeting regime. Box4. Pre- conditions of inflation targeting. Literature on inflation targeting recommends a series of pre-conditions that must be met in order before implementing this regime. Eichengreen et al. (1999) sees lack of the technical capacities of central banks as a fundamental obstacle for the implementation of inflation targeting. Other scholars have maintained a balance in this direction. While listing a number of pre-conditions that must be met, Carareet al. (2000) remains faithful to the idea that such “pre-conditions” must not be considered as “strict pre-conditions” for starting to implement the inflation targeting. This means that a central bank may proceed with its implementation even if some of these pre-conditions have yet to be fulfilled. Nevertheless, they reiterate the need for required policies aiming to satisfy these pre-conditions in the future. Jonas and Mishkin (2000) share a similar view. IMF (2005) highlights that none of the transition countries met all pre-conditions before proceeding with the implementation of the inflation targeting. In World Economic Outlook, September, 2005 IMF divides the pre-conditions in four categories: • Institutional independence. • Developed technical infrastructure. • Economic structure. • Sound financial system. With exception of the direct funding of the budget deficit, Bank of Albania envisages that it will meet the pre-conditions in the end of 2008. The issue of debt funding requires an amendment of the legal framework, i.e. of the law “On Bank of Albania” Which in compliance with the National Plan for implementing the Stabilization Association Agreement is expected to be approved no later than end of 2009. The new law must be completely in line with EU acquis communitaire, thus making compulsory the requirement to prohibit deficit funding from the central bank. This moment will mark the official transfer to a complete inflation targeting regime in Albania. The central bank has compiled during all these years, a list of technical problems that need to be tackled, aiming at improving not only the understanding of inflation sources but also the monetary policy transmission mechanism and of functioning of the Albanian economy as a whole in order to come up with appropriate responses to different shocks that might hit the economy and prices. This list consists in: (i) building up forecasting models for inflation; (ii) evaluation and developing the inflation forecasts models; (iii) initiating a macro-economic model for Albania, more simple modes of inflation that could be used alternatively or complementary to forecast inflation, like VAR et.; (iv) improving the statistical framework, especially the short term real sector data. After 2005, BoA changed the objective form from the interval form of the objective of 2-4 per cent to an objective of 3 per cent in a tolerance margin of +/-1 per cent. Furthermore, BoA reviewed the time span of the inflation objective from a one - year framework fixed at the end of the year to a continual horizon. Such a change was made public and explained in the Monetary Policy Paper of 2006 and was extended to a three - year period of time in the “Development of BoA during 2006 – 2008”. The new form of the objective aimed to enable the focus of inflationary expectations and their anchorage to a single value of 3 per cent. This supposition stemmed from belief that in the interval form of the objective, inflationary expectations are generally oriented towards the upper limit of the target interval. Currently BoA is using a portfolio of the models for inflation forecasting. BoA personnel and the decision – making body (Supervisory Council) are aware of these models’ limitations that originate from problems related to data, short experience in the modeling process and difficulties in assessing expected developments of exogenous variables. This is the reason why current efforts are focused in two directions: (1) Priority is given to the improvement of existing instruments for forecasting. All the six models are improved and updated continually with new indicators that show a better continuity of the explanation and forecasting of inflation. Currently, the forecasts provided by the models over the past 18 months are being re-assessed (Kota et al. 2006). In 2007, the continuity of each model was assessed formally and inflation forecasts were selected. The results showed that quarterly models of forecasting provided better results than the monthly models, while the quarterly database is more complete and more reliable (Kota et al. 2006). BoA maintains that in the near future the monthly models will be replaced by the quarterly models considering that the focus of the monetary policy of B0A will move on to three months timeframes. (2) The second focus of efforts is the development of a macroeconomic model for Albania (meam). Dushku (et al. 2006) makes a technical description of the model. Although progress is observed in this direction, ‘meam” is a continual process that consists in the upgrade of statistical data base, in developing equations and in strengthening relations between the structure of the model and the monetary policy. Results of “meam” are intended to be used for providing reliable macroeconomic forecasts in 2008, in the earliest variant possible for retrieving reliable results. During the period prior to retrieving such results, “meam” will be used for simulating macroeconomic shocks, and this will be of assistance to the decision making of BoA with regard to the monetary policy. Policy Implementation Positive macroeconomic environment and low and stable inflation rate enabled to BoA to gradually reduce the monetary conditions in the economy starting from 2000. Such a policy relieved overappreciating pressures towards the exchange rate and made room for extending financial mediation. Nevertheless, loans mediations occurs mainly for loans in foreign currency and the low interest rate in international markets encourages this phenomenon. With exception of the banking sector, the financial sector remains still undeveloped and includes only some small institutions involving insurance and pensions but lacks of stock market. Having a clear understanding of transmission of monetary policy it is not an easy task for an emerging market economy characterized by low financial markets and subjects to frequent structural shocks. Investigating all the possible channels through which monetary policy could affect inflation: interest rates channel- money view; exchange rate channel; bank lending channel- credit view; inflation expectations channel; proved a very useful exercise to get some good indications on what are the most appropriate instruments to achieve the inflation targeting in Albania: money supply, short – term interest rates, exchange rate etc., or all combination of them. Inflation Targeting (IT) and exchange rate Choosing a monetary regime based on a strict regime of the exchange rate would require appreciation of the real exchange rate. In such a situation, the economic competition with foreigners would be weakened and trust of the central bank would be damaged for fulfilling the objective if the effect would be reduced through a nominal depreciation of the exchange rate. Furthermore, such a regime made it difficult to reduce specific challenges of the transition countries such as the high capital flow in the form of foreign currency remittances. Underlining these issues, IMF (1997) suggested to the transition countries with a regime of fixed exchange rate to transfer to a regime of floating exchange rate after reaching a low inflation rate. Now days, in terms of exchange rate policy, Albania could be better defined under floating regime with occasional interventions in the forex market to maintain a certain level of international reserves. Although the Exchange rate pass-through is quite high especially during depreciation periods, BoA so far has been unwilling to rely on direct forex interventions to meet its inflation targets. This is partly because of changes in fundamentals in periods of strong depreciations (like 1997) and the insufficient level of international reserves (the level of it is fixed on an annual basis, in cooperation with WB and IMF, and nowadays the floor is under the value of 4 months imports- actually at EUR 1.5 billion). From 2001 Exchange Rate has shown noticeable stability (Fig.9). The ER pass-through effects have also dropped (Fig.10). Fig.9 NEER & REER2) of Lek vs.80% Euro and Fig.10 % of ER fluctuations of less than 1% 20% USD in Albania within 2 year period Running a flexible monetary policy like Monetary Target and Inflation Target does not mean that monetary policy has to disregards the exchange rate. Monetary policy under Inflation Target regime affects exchange rate channels in several ways. There are two main reasons, I think, why exchange rate channel should be closely watched. (1) Fear of pass- through: while direct interventions have been avoided so far, monetary policy has been particularity attentive to interest rate differentials between Lek and Euro/USD to avoid currency (deposit) substitution which is turn could trigger an inflation-depreciation spiral (97’ shock: BoA increased interest rates by more then 10 percentage points). One issue which requires further attention is interest rate misalignments – can ___________________ 2) NEER is the weighted average of bilateral nominal exchange rates of the home currency in terms of foreign currencies. The REER, defined as a weighted average of nominal exchange rates adjusted for relative price differential between the domestic and foreign countries monetary policy keep its interest rate over the “natural rate of interest” for unlimited time. (2) Fear of floating: although euroisation in Albania can be classified as moderate compared to other countries in region (like Croatia), it still remains significant enough to be disregarded for its inherent risks and implications on economy. In this environment central bank becomes rather sensitive to exchange rate volatility even under a free float. A high degree of foreign currency substitution (euroisation, dollarisation) can cause serious problems for Inflation Targeting (Mishkin, 2000). Partial euro/dollarisation has the potential to make inflation targeting regime, which requires some degree of exchange rate flexibility, vulnerable to financial instability (Mishkin and Sevastano, 2002). In Albania the level of euro/dollarisation in terms of foreign deposits to total baking assets is around 20-36 percent. The problems related to euro/dollarisation need to be treated with caution in the prospect of the declining foreign remittances that so far has reduced the impact of drop of current account deficit and have ensured a relatively stable exchange rate. Unilateral euro/dollarisation in countries like Albania is already taking place not simply because of monetary policy (or several others) differences as much as due to the tight links these countries have with EU (70+ percent of Albanian import/export with Italy and Greece, 30 percent of labor force is established to be working abroad, mainly in these countries). To ensure some flexibility these countries have let open the option of euroisation (holding fx deposits/loans, and carrying out transition in euros). Therefore a partial euroisation unintentionally is already taking place. In Albania e.g. quite 80% of housing transactions are done in euro. Measures to “stop” euroisation may not be that different from measures to reduce cash economy. Here we could mention the enforcement of transactions in domestic currency especially of durable goods and long term contracts and of loans. Still the question remains: how promising is this and is it worth pursuing? If yes, question rise: could exchange rate serve as an extra instrument to central banks beside interest rates particularly concerning for forthcoming capital inflows? The uncertainties about monetary policy transmission mechanism make it difficult to foresee the long indirect effect of interest rates and exchange rates on inflation (Orr et al.98). While exchange rates and interest rates could be complementary tools in short run it is less likely so in the long run, especially when the relative importance of real shocks is greater. IT and Interest and Banking Channels A sound financial system is important to the success of IT for several reasons. A financial system could better sustain the negative impact of eventual exchange rate volatility related to the high dollarisation. Also, considering the several objectives BoA is following, it might be important to examine whether Albania has missed or could miss the inflation target if it conflicts with reaching other objectives such as financial stability. A sound financial system would put BoA under less pressure to trade off between objectives. The development of the financial system is an important factor also for the conduct of monetary policy. The actual level of financial development in Albania is far from being adequate for an effective monetary policy. The level of competition among banks at the moment is still low while the money and stock markets are virtually non existent. As mentioned in the fist part of the draft, a turning point was marked by the successful privatization of the Savings Bank of Albania (now Reiffaisen Bank), by far the largest bank of the system, which until the end of 2003 was state-owned (Box.3). This privatization sparked a series of positive changes in the banking system such as strong credit growth for example. In 2005 the growth of credit reached an extraordinary velocity of almost 82 percent, though from a low base. Box. 4 Albanian Banking System-Main features Today, in Albania act 14 foreign – owned banks, out of 17. Some big international names entered Albanian banking system in 2007: Intesa SanPaolo IMI bought American Bank of Albania and Italian-Albanian Bank (both banks were merged and by first of January 2008, operate under American Bank of Albania trademark); Société Générale bought 75.01% of Albanian Popular Bank. A highly concentrated banking system emerges from statistics, where top four banks have a share of ~ 73 percent of total assets. There is no public (stateowned) bank in Albania any more; The Albanian Government owns only a 40% equity stake at United Bank of Albania (~EUR 3.7 mio). Credit of householders has grown proportionally with total credit growth with small advances against credit to business (Fig.11). But credits to business constitute the lion share of total credit. A considerable share of credit is extended in foreign currency. Credit in Lek (local currency) has gained some grounds recently, mainly due to the fatigue of funds (deposits) in foreign currency, but the preferences of banks and businesses remain influenced toward foreign exchange credit. Fig.11 Household credit against total volume of credit (in mln of lek) The long period of macroeconomic stability has created kindly monetary conditions, reflected in low or declining interest rates and a stable or appreciating exchange rate, even the credit ratios still low compared to EU and SEE standards (Fig.12). However, monetary policy transmission mechanism becomes more uncertain as financial markets get deeper. The main concern relates to the interest rate channel effectiveness. Previously this channel has been adequate to influence the exchange rate by affecting prices of traded goods. However, as price pressures from the rapid credit 6000 50 5000 40 4000 30 in % in mln EUR Credit/Deposit ratio 2001 - 2007 3000 20 2000 10 1000 0 2001 2002 2003 Credit to Economy (mln EUR) 0 2004 2005 Deposits 2006 2007 Credit/deposit ratio Fig.12 Credit/Deposit ratio 2001-‘07 growth increase, interest rates channel becomes weaker. The growth rate of credit is relatively unresponsive to interest rate changes, given the fact that loans are largely extended in foreign currency. Tightening of monetary policy would simply lead to appreciation with little impact on credit volume. Forex intervention to avoid excessive appreciation needs sterilization to maintain monetary policy tightness. The recent experience though, has shown that credit channel may not be completely lost. Although, local interest rates play little role in influencing the demand of foreign currency loans it may restrict their supply through deposits’ rates differentials. If central banks increase local interest rates, besides restricting demand for domestic currency loans, reduces also the incentives of savers (investors) to put their money in foreign exchange deposits which should reduce banks’ funds for extending foreign currency loans. There are two “get-outs” with the above transmission channels: First, foreign banks may force their own interest rate differential that would dominate central bank interest rate policy; Second, even if private banks follow central bank interests they may borrow some foreign exchange resources from abroad, which would be a typical case of credit boom. If the above transmission channel fails, then central bank is left with direct instruments such as credit ceiling. However, some other supervisory measures have been introduced as an alternative to slow down growth before recurring to credit ceilings, which at the same time will aim to contain weakening of banks’ loan portfolio. Some of these measures are: increasing reserve requirements on foreign currency deposits above 10 percent; capital adequacy will be amended to allow risk weights up to 150 percent to apply to high risk loans or portions of loans; impose higher then 12 percent minimum capital adequacy requirements on individual, relatively high-risk, banks; and introducing stricter requirements for collateral/loan value ratios. The importance of defining the target From the start of the transition process the Bank of Albania was given the task of reducing and keeping inflation at low levels. However, as in many other independent central banks’ laws, the Bank of Albania Law does not specify precisely what low inflation is. Even if the ongoing debate on the right degree of flexibility in defining the exact level of the target, a consensus has been emerging among academics and central bankers that low inflation should be close to zero but positive3). Some central banks in particular those applying inflation ______________________ 3) Inflation rates slightly higher then zero are usually sustained on the bases of: first, once inflation gets negative, which is very likely in a strict zero target, it may be very difficult to get out of deflationary cycle; second, at inflation rates close to zero, rigidities could make it costly to sustain further inflation reductions; third, possible measurement biases in CPI asks for some allowances when targeting zero inflation targeting like European Central Bank (ECB), have defined the inflation level they target, while others like FED have been more reluctant to make public a precise number. While this regime does not necessarily require announcing an explicit inflation target, in the case of Albania there has been a visible objective for inflation too. Thus, from the first Enhance Structural Adjustment Facility (ESAF) Program signed with IMF in 1993 it was clearly stated: Following the near-slide into hyperinflation in 1991-92, the reduction of inflation to about 15 percent by 1996 is a key program objective: (IMF, 1993para.12). Inflation objectives were revised each year and were regularly mentioned in Bank of Albania annual Reports, alongside monetary targets, which has not been a common practice in a pure monetary targeting regime (e.g. Bundesbank). The way inflation objective is announced or defined has been adjusted to take into account the above problem. The definition of meeting end-year inflation (Dec. to Dec) of 2-4 percent has been replaced in bank’s reports with meeting annual inflation of 2-4 percent in mid-term. Mid-term horizon at the moment is defined at around one year. The exact length of the horizon could be extended depending on many factors which are being analyzed continuously ranging from monetary policy transmission mechanism time length to the confidence with which BoA assesses future inflationary pressures. For purely transparency reasons inflation definition has remained headline inflation despite the recent decision to introduce a core inflation definition by INSTAT. Switching to core inflation which excludes food and energy items- subject to strong seasonal and weather changes largely beyond monetary policy range of influence – would simply disconnect the bank from public. To that level achieving and maintaining a target of 2 to 4 percent for mid - term period is still an appropriate objective for Albania for several reasons: (1) Large share of food in the consumer basket; (2) Historical reasons; (3) Balassa-Samuelson effects; (4) Imported Inflation. (1) Large food weight As mentioned above the large weight food consumption in Albania increases the volatility of prices. One way of getting rid of seasonal volatility is to target average inflation. Other types of non – seasonal volatilities should be addressed by widening the band. The 2 to 4 band in the last 5 years have proved to be reasonably appropriate to account price volatility in spite of the changeable weather during this period. This brings us to the historical reasons. (2) Historical reasons Albania has shown remarkable results in term of price stability. Inflation has been distinguished generally by lower rates relative to other transition economies and not far from inflation seen across EU countries. After the initial period of transition 1991-1992 that saw inflation levels as high as 200 percent, single digit inflation rates prevailed until second half of 1996 when it started picking up over 20percent, largely due to fiscal dominance that characterized that year. In 1997 the pyramidal schemes events triggered an economic and social chaos bringing the economy to a standstill and boosting inflation up to 50 percent. After the turmoil of 1997, inflation has been brought again under control to about 2-4 percent 1999, where it has stabilized ever since. Both disinflation periods at the beginning of transition and after 1997 have been short, without any obvious real production and employment costs. The actual band of 2 to 4 where inflation seems to stabilize when the economy is not subject of major shocks could be interpreted as equilibrium that for a mid-term period appears to be sustainable. Nevertheless, this is a dynamic and fragile equilibrium. The catching up process of the economy may have several implementations for inflation. An often mention implication takes place through Balassa – Samuelson (BS) effect. (3) Balassa – Samuelson effect Productivity convergence constitutes important adjustment that may have significant implications for inflation in emerging markets. So far BS effect does not appear to have been a significant inflationary source in Albania. However, possible effects of BS on the exchege rate have to be taken into consideration. Olters (2005) offered empiric evidences which have shown that the presence of this effect has an important significance. Whether this will change in the future will very much depend on the intensity of the forthcoming structural change the economy will undergo. If any of these developments occurs, most probably it will require upward revisions of the band to accommodate for these necessary real adjustments. (4)Imported Inflation Inflation can be divided into two components, one that reflects domestically generated inflation and one that reflects imported inflation: Total inflation = weight x domestic inflation + (1-weight) x imported inflation Where: weight is domestic inflation’s share of total inflation and in Albania is estimated to be 56 percent, which means that the weight of imported inflation component is 44 percent (source: INSTAT). The above three points were mostly related to domestic inflation. But, Albania is a small and open economy with a large share of imports 42 percent of GDP). This means that imported inflation could be quite significant. Unlike previous factors imported inflation is a component that most likely pushes inflation down toward 2 percent level bearing in mind that Albania’s two main trade partners are Italy (with 32 percent of total imports) and Greece (16 percent) both Eurozone members that targets below 2 percent inflation rate. A final argument, which is more, comforting that convincing, is the experience of other countries in targeting inflation. As table below shows inflation targets across these countries vary from 1 to 6 percent, with point targets on average 2 percent for industrial countries and about 3 percent for emerging countries. The target range is set about 1.0 to 1 .5 percentage points around the point target for most Inflation Targeters. Apart from the countries that at the time of Inflation Targeting implementation have had high inflation, the range of the inflation target is on average 2 percentage points. Many of countries that apply wider bands tend to narrow them as inflation gets closer to their long – term inflation objectives (Tab.3). Table 3 Source IMF Emerging Markets Target Israel 2.0 Czech Rep 3.0 Korea 3.0 Poland 2.5 Brazil 4.5 Chile 3.0 Colombia 5.0 S. Africa 4.5 Thailand 1.7 Mexico 3.0 Hungary 3.5 Peru 2.5 Philippines 4.5 Band 1-3 ±1.0 2.5-3.5 ±1.0 ±2.5 1-4 ±0.5 3-6 0-3.5 ±1.0 ±1.0 ±1.0 3-6 Industrial Countries Target N. Zealand 2.0 Canada 2.0 UK 2.0 Australia 2.5 Sweden 2.0 Switzerland 2.0 Iceland 2.5 Norway 2.5 Band 1-3 1-3 2-3 ±1.0 <2.0 A final point should be 3 ±1% or simply a range of 2 to 4 percent. Again this depends very much on the confidence with which BoA controls inflation. The former type of definition helps anchoring inflation expectations better. However, it may also put more pressure on central bank to explain inflation being below or above the centre target even if it is within the ±1% range. In other words it may behave and have the same implications as a point target. As I mentioned above, BoA from 2001 has been announcing the 3 ±1% type of objective rather than 2 to 4 range. So far, it seems it has been a rewarding asymmetrically, especially from above, in case real economic adjustments will require higher inflation rates. ****** BoA aims to implement the inflation targeting as an efficient regime for the implementation of the monetary policy. The implementation of such regime is motivated by a number of reasons. Inflation targeting encourages a higher level of transparency for the monetary policy and as such, this regime provides the central bank with a unique means for communicating the monetary policy to the public. This peculiarity allows a more effective administration of the inflation expectations, reduction of the risk prime as well as a better continuity for the fulfillment of this objective. Furthermore, the implementation of the monetary policy under inflation targeting is established on a wide base of information, thus giving to BoA the opportunity to include in the decision-making structure the increasing complexity of the Albanian economy. This allows to the central bank the required flexibility towards possible macroeconomic shocks. Inflation targeting, defined as a regime of “limited freedom” is expected to make possible a better balance regarding the monetary policies: determining the necessary limitations in order to avoid time discrepancy of the monetary policy – represented by increased reliability – as well as the flexibility required for administrating specific shock factors of the economy. Considering that inflation targeting is accompanied with a regime of fluctuating exchange rate, it provides room for efficient administration of external shock factors. Lastly, benefits that BoA will derive from capacity building for the implementation of inflation targeting should not be considered as short - term benefits. In the long term, the way of monetary policy is entertained with the adaptation of the Euro, a long process if considered from the number of years that it would take for this process to be concluded. Conclusions The nominal anchor used by the Bank of Albania in the last decade to control inflation has been money supply. Although not regarded as an orthodox regime nowadays as it used to be some twenty years ago, monetary targeting has served Albania well judging from the low inflation record. But the instability of money demand increases as advances and structural changes of the financial systems take place. The recent wake up of banking sector in Albania has urged BoA to reconsider the actual regime, as it may become inappropriate in attaching inflation expectations effectively in the future. Inflation Targeting was considered as the most appropriate alternative. Although some doubts have been raised about its applicability in developing countries it is a good framework for at least making clearer Central Bank’s intentions to maintain low inflation, while allowing for some flexibility to real shocks (a major advantage over exchange rate regimes). Several prerequisites of inflation targeting are already present in Albania: (a) independent central bank targeting inflation, (b) instrument independence, (c) free floating exchange rate. Explicit inflation targets have been announced since 1993 besides the monetary targets. Some other elements are been developed further: (i) forecast models, (ii) monetary policy (inflation) report. However, other preconditions like well understood and effective interest rate transmission mechanism, fiscal dominance (potentially) prove to be more challenging to deal with. Most of these inflation targeting preconditions however are indispensable for the success of any other regime too, including the actual money targeting (MT). Therefore, inflation targeting adoption cannot be simply judged based on a standard compliance to these criteria. *** Recommendations Two critical problems need to be tackled for accelerating the progress: (1) the inadequate technical expertise to build up the necessary tools of inflation targeting implementation, and (2) the poor state of economic data. The former has been addressed through technical expertise from international institution like IMF. But, to make technical progress sustainable over time, a strategy for developing inhouse human capital need also to be developed further. Statistical issues represent one of the most challenging areas for starting to implement inflation targeting. Regarding statistical data drawbacks, several steps need to be taken to improve data sources such as: (a) carrying out surveys, interviews, (b) assisting INSTAT to improve national account statistics frequency and quality, and out of government politics as well. (c) determining new indicators for prices of immovable assets, annual index of industrial production and foreign debt on a quarterly basis, surveys on foreign direct investments and measuring remittances (d) reducing time for collecting and processing monetary data. references Bank of Albania, Annual Reports. Bolle, M., Meyer, T., (2005). “The efficiency of monetary policy in Albania and need for further reforms”. “Central Banking in integration period”, pp. 19-52. Carare, A., Schaechter, A., Stone, M., Zelmer, M., (2002). “Establishing initial conditions in support of inflation targeting”. Imf -Working Paper. Çeliku, E., Shtylla, D., Hashorva, G., Hoxholli, R., Kota, V., (2006). “Economic models portfolio in forecasting inflation” BoA - Working Paper Dushku E., Kota V., Binaj G., (2006). “ A macroeconomic Model for Albania”Boa- Working Paper. Estella, A. and Mishkin, F. (1997) “Is there a role for onetary aggregates in the conduct of monetary policy?” Journal of Monetary Economics, Vol. 40, pp, 279304 Eichengreen, B., Masson, P., Savastano, M., Sharma, S., (1999), “Transition strategies and nominal anchors on the road to greater exchange-rate flexibility,” essays in International finance, no. 213. IMF, (1997). “Monetary and financial policies in transition countries” World economic outlook, pp. 98-118. IMF, (2005). “World economic outlook” Sept. 2005. pp. 172-178. IMF, (2005). “Albania – Imf country report” Jan. 2006. Jonas, J., Mishkin, F., (2003). “Inflation targeting in transition countries: experience and prospects”, Working Paper no. 9667. Kota, V., Hashorva, G., Peeters, M., Çeliku, E., (2006). “ Evaluation of target inflation forecast, April 2005 -September 2006”. BoA WP Luçi, E., Ibrahimi, F., (2005) “Monetary targeting regime in Albania and some future references”; Masson, P.R. Savestano, M.A. and Sharma, S. (1997) “The scope of Inflation Targeting in developing countries”, IMF working paper no. WP/97/130 Mishkin, F. (2000) “Inflation targeting in emerging-market countries” American Economic Review, Paper and proceedings, vol. 90, no.2, pp. 105-109; Mishkin, F. and Sevastano, M.A. (2002) “Monetary Policy Strategies for Emerging Market Countries: lessons from Latin America” vol.54, no.2, pp.45-82 Mishkin, F.S. (2002) “Monetary Policy”, NBER Reporter Winter. Sojli, E. (2004) “An evaluation of currency substitution in the Albanian economy”, forthcoming, BoA, WP Stone, M.R. (2003) “Inflation Targeting Lite”, IMF WP no. WP/03/12 Svenson, L. (1997) “Inflation Forecast targeting-Implementing and Monitoring Inflation targets” NBER WP no.5797. Olters,J-P.,(2005).“Is Inflation in Albania to low? -Macroeconomic fundamentals and social-economic development” pp 147-174. Orr, A., Fischer, A. (1998)“Asset prices and monetary policy”-Working paper Shtylla, D., Suljoti, E., (2006) “ The transition from monetary policy report to inflation report of BoA” Bank of Albania -WP