Governor Mugur Isărescu’s intervention in
Session 1: Monetary and exchange rate policies in CESEE countries – experience gained with different frameworks and current challenges

Ladies and Gentlemen,

It is an honour for me to be invited to share in front of such a distinguished audience the NBR’s experience with the inflation targeting regime, including the context and reasons for its adoption. After more than a decade of monetary targeting, the NBR joined the club of inflation targeters in August 2005. I could say that this conference on monetary policy in CESEE countries comes at the right time to look back at a decade of inflation targeting in Romania. All the more so that it has, by no means, been an ordinary decade – (i) economic boom and widening macroeconomic imbalances amid large capital inflows until 2008, (ii) the subsequent bust following the outbreak of the global economic and financial crisis, with the whole plethora of problems related to correcting internal and external imbalances, (iii) the recent creditless recovery, amid an unusual concurrence of favourable supply-side shocks.

But let us start with the beginning of monetary policy in Romania. At the time when the National Bank of Romania had to choose a strategy for conducting its monetary policy, i.e. in the early nineties, monetary targeting was actually the only feasible option. Not only the low forex reserves precluded a fixed exchange rate arrangement, but, even if possible, the alternative was not suitable given the widening external imbalance.

However, by the end of the decade, monetary targeting had reached its operational limits given the emergence of a break in the relation between monetary aggregates and inflation, a phenomenon which matched developments elsewhere in the world. In the case of Romania – and probably in the case of other transition economies, as well – a reason for the monetary aggregates unsuitability as intermediate target was attributable to the lack of financial discipline which led to the build-up of arrears in the inefficient SOEs (largely, towards the state and among themselves). As the arrears acted as money substitute, the role of monetary aggregates in guiding monetary policy became increasingly irrelevant – restricting money supply often resulted into an accumulation of arrears.

Faced with the need to choose an alternative monetary policy strategy, Romania opted for inflation targeting as early as 2002, even though it took several years of preparations until the actual adoption, in August 2005. We also took into account pegging the exchange rate, as other countries in transition did, such as Hungary, which in the first half of the ’90s implemented the crawling peg, or Bulgaria, which adopted the currency board while facing a major financial crisis. However, neither solution has been deemed appropriate for Romania in the pre-accession period. The crawling peg helps control inflation when it is brought down from 30-40 percent, but it is less useful when a lower level is envisaged, since the crawl incorporates inflation expectations. Moreover, exchange rate flexibility was seen as an asset for an economy as big and rigid as Romania was at the time, especially given the upcoming liberalisation of capital flows.

When introducing inflation targeting, the NBR opted for the “light” version of the strategy, which meant retaining the managed float feature of the exchange rate regime. The actual form inflation targeting took in Romania thus matched the “managed-floating plus” concept introduced by Goldstein in 2002, considered the most suitable choice for an emerging-market economy involved in the global capital market, as it combines the managed float part, allowing for FX market interventions in order to smooth out excessive volatility, with an inflation targeting monetary policy strategy and an active pursuit of measures to limit the degree of currency mismatches in the economy.

The first two years of inflation targeting overlapped the closing stages of capital account liberalization and I would dare say the process went a little bit faster than advisable. We learnt before long how challenging it is for monetary policy to manage aggregate demand and expectations in a small open economy amid a flood of capital, particularly when having to cover also for a strongly pro-cyclical fiscal policy.

Naturally, monetary policy embarked on a countercyclical path, but the policy rate was not enough to fight the credit boom effectively, especially since it developed along the foreign currency component. The latter phenomenon was, of course, associated with the massive presence of foreign-owned banks, largely as a result of privatizations undertaken with major European groups (SocGen, Raiffeisen, Erste). After a string of domestic bank failures (particularly in the late nineties), such a development emerged as a natural solution for introducing stronger corporate governance and modern banking practices.

In such a context, the interest rate hikes required in order to manage aggregate demand and anchor expectations would have entailed further capital inflows and an unsustainable nominal appreciation of the exchange rate. Initially, the NBR made full use of the reserve requirement ratios, which were increased up to 20 and 40 percent for lei- and forex-denominated liabilities respectively in 2006. With reserve requirement ratios already at levels well above those applied by any other central bank in the EU and with annual credit growth rates still above 50 percent, the NBR had no choice but to innovate.

Apart from moral suasion, a number of administrative measures – controversial at the time, but now considered valid macroprudential instruments – were added to its toolbox, such as: enforcing maximum loan-to-value ratio, introducing debt service ceilings relative to households’ monthly disposable income (the scope of which was subsequently expanded to include non-bank financial institutions engaged in lending), setting limits to banks’ forex exposure vis-à-vis unhedged borrowers, using differentiated coefficients in stress tests (higher for exposures to EUR than lei, with even stricter coefficients for CHF- and USD-denominated credit).

The countercyclical stance during the boom years allowed monetary policy to turn stimulative after the outbreak of the global crisis: the interest rate was gradually cut from 10.25 percent at the beginning of 2009 to 1.75 percent currently, while the reserve requirement ratios were lowered to 8 and 14 percent for lei- and forex-denominated liabilities respectively.

This conduct, together with the switch in liquidity conditions, allowed the central bank to make up for the losses accumulated during the boom years as a result of sterilization operations. While profitability is not a concern for a central bank, this development proved fortunate from the perspective of the political economy. It would have been inappropriate for the central bank to contribute to the quasi-fiscal deficit at a time when the government was engaged in a massive consolidation effort. Moreover, it also helped credibility, considering that commercial banks were subject to regulatory calls for additional capital.

As I mentioned before, in the case of Romania, monetary policy did not have to navigate in the uncharted waters of quantitative easing and negative nominal interest rates, but nevertheless it is not bereft of challenges.

I would list first the difficulty in designing monetary policy in a context where the negative output gap is closing fast, while a perfect storm of favourable supply-side shocks – namely oil and food price drops and indirect tax cuts – has been generating low inflation readings for quite a while now, with the potential to alter inflation expectations in the medium term.

Another challenge is the fact that there are still disruptions in the transmission mechanism of the monetary policy. Specifically, even though the interest rates on new loans in domestic currency have declined further in response to the policy rate cuts, the annual growth of credit to the private sector has leveled off in recent months.

While this was mainly the outcome of the ongoing banking system balance sheet clean-up, it also reflected the impact of a number of factors that are hampering the functioning of the monetary policy transmission mechanism (in particular that of the risk-taking channel). The still high, albeit significantly declining, level of NPLs, the persistently elevated risk aversion of banks, and the deleveraging of euro area credit institutions are such factors that further depress credit supply. At the same time, the insufficient gains in confidence, particularly in the case of businesses, hinder credit demand.

On the bright side with reference to the transmission mechanism, I should point out the steady improvement testified by the ever expanding share of local currency-denominated credit in total lending, which has recently increased to 45 percent and looks set to reach 50 percent by the end of the year.

The quest for deterring speculative capital inflows and cooling off foreign currency lending stood behind some peculiarities of the operational framework of monetary policy in Romania. I am referring here to the relatively large width of the interest rate corridor and the unusually high reserve requirement ratios. In the past two years, the NBR has capitalized on the window of opportunity provided by the favourable macroeconomic and financial background to take steps towards rectifying the situation, yet the process needs to be continued until reaching the prevailing levels across the EU.

Looking ahead, it is essential that the NBR identify, as it has until now, the right timing for the upcoming stages of the alignment process to unfold without impairing the smooth functioning of the markets.

This issue of normalizing the operational framework has a particular relevance in the context of Romania’s plans to adopt the single currency, as the adjustment should be completed before the ERM II entry. The horizon for euro adoption should therefore allow for enough time to carry out this adjustment, as well as to achieve sufficient progress in real convergence. Even if the Maastricht criteria are currently fulfilled, the sustainability of this achievement can be ensured only with these prerequisites in place.

I will end my remarks here, hoping that at least some of the issues I have tackled prove useful to conference participants.