Russia’s real GDP fell by a modest 3 percent in 2020. The question addressed here is how a major oil-exporting country can go through the COVID-19 pandemic with a decline of this magnitude when oil prices fell by 35 percent at the same time as the domestic economy suffered from lock-downs. The short answer is that it is mainly a statistical mirage. The aggregate real GDP decline would have been almost three times greater than in the official statistics if changes in exports were computed in a way that better reflects their value. In particular, the real GDP calculation uses changes in volumes rather than values to omit inflation, but for exports, it thus ignores large changes in international oil prices. In the end, what the government, companies, and people in Russia can spend is much more closely related to how much money is earned on its exports than how many barrels of oil the country has sold to the rest of the world. More generally, this means that real GDP growth in Russia is not a very useful statistic in years with large changes in oil prices, as was the case in 2020, since it does not properly reflect changes in real income or spending power. When policymakers, journalists, and scholars now start to compare economic developments across countries in the covid-19 pandemic, this is something to bear in mind.
The world is closing the books on 2020 and it is time to take stock of the damage done by the COVID-19 pandemic thus far. A year into the pandemic, over 100 million cases have been confirmed and almost 2.5 million people have died worldwide according to ECDC (2021) statistics. Russia has not been spared and Rosstat reported 4 million infected and over 160 000 dead in 2020.
Human suffering in terms of lost health and lives is certainly the main concern in the pandemic, but on top of that comes the damage done to economies around the world. Falling incomes, lost jobs, closed businesses, and sub-par schooling will create significant health and other problems even in a fully vaccinated world for years to come.
Understanding how real GDP has fared in the crisis does not capture all of these aspects, but some. With the IMF’s latest World Economic Outlook update on economic performance out in January 2021, it is easy to start comparing GDP growth across countries (IMF, 2021). GDP growth is a standard measure of past performances in general, but the numbers for 2020 may also enter various domestic and international policy discussions of what does and does not work in protecting economies in the pandemic. For countries that seem to have fared better than their peers, the growth numbers are likely going to be used by incumbent politicians to boost their ratings or by consumers and business leaders making plans for the future.
In short, real GDP numbers are important to most economic and political actors, domestically and globally, with or without a crisis unfolding. It is therefore important to understand how Russia, a major oil exporter with significant losses of lives and incomes in the pandemic, could report a real GDP decline of only 3 percent in 2020 (Rosstat, 2021). Although this is not far from the global average reported by the IMF (2021), it is far better than the 7.2 percent drop in the Euro area, 10 percent fall in the UK, or 7.5 to 8 percent declines of its BRICS peers, South Africa and India. This brief provides the details to understand that Russia’s performance is more of a statistical artifact than a fundamental reflection of the health of the Russian economy.
Oil prices, GDP growth, and the ruble
Russia’s dependence on exporting oil and other natural resources is well documented (see for example Becker, 2016a and 2016b) and often discussed by Russian policymakers and pundits. In particular, changing international oil prices is a key determinant of growth in the Russian economy. Even if the level of real GDP disconnected from oil prices somewhere between 2009 and 2014 (Figure 1), the link between real GDP growth and changes in oil prices persists (Figure 2).
Figure 1. Russia real GDP and oil prices
The empirical regularity that still holds is that, on average, a 10 percent increase (decline) in oil prices leads to around 1.4 percent real GDP growth (fall), see Becker (2016a). With a 35 percent decline in oil prices in 2020, this alone would lead to a drop in GDP of around 5 percent.
Figure 2. GDP growth and oil price changes
One factor that has a fundamental impact on how the relationship between oil prices and different measures of GDP changes over time is the ruble exchange rate. For a long period, Russia had a fixed exchange rate regime with only occasional adjustments of the rate. A stable exchange rate was the nominal anchor that should instill confidence among consumers and investors. However, when changes in the oil prices were too significant, the exchange rate had to be adjusted to avoid a complete loss of foreign exchange reserves. This was evident in the 90’s with the crisis in 1998 and later in the global financial crisis in 2008/09. Eventually, this led to a flexible exchange rate regime and in 2014, Russia introduced a flexible exchange rate regime together with inflation targeting as many other countries had done before it.
As can be seen in Figure 3, this has important implications for how changes in international oil prices in dollars are translated into rubles. Note that the figure shows index values of the series that are set to 100 in the year 2000 so that values indicate changes from this initial level. Starting in 2011, but more prominently since 2014, the oil price in rubles has been at a significantly higher level compared to the oil price measured in dollars, which is of course due to the ruble depreciating. This affects the government’s budget as well as different measures of income in rubles. However, if oil prices in dollars change, this affects the real spending power of Russian entities compared with economic actors in other countries regardless of the exchange rate regime. Moving to a flexible exchange rate regime was inevitable and the right policy to ensure macroeconomic stability in Russia when oil prices went into free fall. Nevertheless, it does not change the fundamental economic fact that falling oil prices affect the real income of an oil-exporting country. It also makes it even more important to understand how real GDP is calculated.
Figure 3. Oil prices and exchange rate indices
The components of real GPD
GDP is an aggregate number that can be calculated from the income or expenditure side. The focus in this brief is on the expenditure side of GDP. The accounting identity at play is then that GDP is equal to private consumption plus government consumption plus investments (that can be divided into fixed capital investments plus change in inventories) plus exports minus imports (where exports minus imports is also called net exports). Being an accounting identity, it should add up perfectly but in the real world, components on both the income and expenditure sides are estimated and things do not always add up as expected. This generates a statistical discrepancy in empirical data.
Another important note on real GDP (rather than nominal GDP measured in current rubles) is that the focus is on how quantities change rather than prices or ruble values. The idea is of course to get rid of inflation and focus on, for example, how many refrigerators are consumed this year compared to last year and not if the price of refrigerators went up or down. This may sound obvious, but it comes with its own problems concerning implementation and interpretations. For Russia, real GDP becomes problematic because its main export is oil (gas and its related products). The price of oil is just one of many drivers of Russia’s inflation but is an extremely important driver of its export revenues and growth as has been discussed above. On top of that, oil prices are volatile and basically impossible to control for Russia or even the OPEC.
So why does this matter for understanding Russia’s real GDP growth in 2020? The answer lies in how the different components of real GDP are computed. To make this clear, the evolution of the components between 2019 and 2020 is shown in Table 1.
Table 1. Russia’s GDP components from the expenditure side
In short, private consumption fell by close to 9 percent in 2020 compared to 2019; government consumption increased by 4 percent; gross fixed capital formation declined by 6 percent while inventories increased by 26 percent; exports lost 5 percent, but imports went down by 14 so that net exports showed an increase of 65 percent! To calculate the impact these changes have on aggregate GDP growth, we need to multiply with the share of GDP for a component to arrive at the impact on GDP growth in the final column of Table 1.
Although there are some issues to resolve with both government consumption and inventory buildup, to understand real GDP growth in 2020, it is crucial to understand what happened to exports and imports in real GDP data. First of all, how does this data compare with the balance of payments data that measures exports and imports in dollar terms or the data that show the value of exports of oil, gas, and related products? Table 2 makes it clear that the numbers do not compare at all! Again, this is due to real GDP numbers being based on changes in volumes rather than values while the trade date reports values in dollars (that can be translated to rubles by using the market exchange rate).
In the real GDP statistics, net exports show growth of 66 percent in 2020, compared to declines of 37 to 44 percent if merchandise trade data is used. Going into more detail, real GDP data has exports declining by 5 percent, while other indicators fall by between 11 and 37 percent. It is similar with imports (that enter the GDP calculation with a negative sign); the import decline recorded in real GDP is 14 percent, while trade data suggest a 6 percent decline in dollar terms but an increase of 7 percent in nominal ruble terms.
Table 2. Trade statistics
What would it mean if we use some of these alternative growth rates for exports and imports (while keeping other components in line with official statistics) to calculate aggregate GDP growth in 2020? The rationale for keeping other components unchanged is that this provides a first-round effect of changing trade numbers on real GDP growth.
To make this calculation, the GDP shares of exports and imports (or net exports) in 2019 are needed. Table 1 shows that these numbers are 27 and 24 percent (or a net 3 percent) of total GDP. Multiplying the share of a GDP component with its growth rate gives the contribution of the component to overall GDP growth. The calculations based on different trade data are shown in Table 3. The last line of the table is what GDP growth would have been with these alternative trade data. Note that the real GDP growth number is -2.9 percent when we use the individual components of GDP decomposition (rather than the official headline number -3.1 real GDP growth when using aggregate GDP) so this is shown here to make the table consistent with the alternative calculations. In the last column of Table 3, oil and gas exports are assumed to make up for half of exports and this number disregards changes in other exports or imports to isolate the effect of changes in the value of oil and gas exports from other changes.
The summary of this exercise is that with more meaningful trade data used in calculating GDP growth, Russia would have recorded a decline of around 9 percent rather than 3 percent. This is of course a partial analysis focusing on the trade part of real GDP since this effect is very striking. Other components of the calculation may also have issues that need to be adjusted to arrive at a more realistic growth number. Still, even the current estimate is not unrealistic. For example, household consumption fell by around 9 percent, which would be consistent with a GDP decline of 9 percent that is not recovered in the future in a permanent income model.
Table 3. GDP growth contributions from alternative trade data
Real GDP growth numbers are important to understand economic developments in a country and provide the foundation for many types of economic decisions. The numbers are also used to compare the economic performance of different countries and evaluate policy responses in the COVID-19 pandemic we are currently part of.
The problem with Russia’s reported growth of minus 3 percent is not that the real GDP calculation is wrong per se, but it is clearly the wrong metrics to use for understanding how incomes and purchasing powers of Russian households, companies, and the government changed in 2020. If we instead use trade data that better reflect plummeting oil prices in international markets, alternative estimates of Russia’s real growth show a GDP decline of (at least) 9 percent. This is a three times larger drop than the official number of minus 3 percent. This is important to keep in mind when Russia’s economic performance in the pandemic is compared with other countries or while discussing the economic realities of people living in Russia.
- Becker, Torbjörn, 2016a. “Russia’s Oil Dependence and the EU”, SITE Working paper 38.
- Becker, Torbjörn, 2016b. “Russia and Oil — Out of Control”, FREE policy brief.
- BOFIT, 2021, Weekly report 7 on Russia.
- Central Bank of Russia, 2021, data on exchange rates.
- ECDC, 2021, data on covid-19 infections and deaths.
- IMF, 2021, World Economic Outlook update, January.
- Roststat, 2021, Russia GDP data.
- U.S. Energy Information Administration, 2021, data on oil prices.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
Recent studies document that emerging markets are rather similar in their growth patterns despite profound differences in starting conditions and productivity fundamentals. This challenges the common view on productivity as the main growth engine. The crucial role of the external environment for emerging markets emphasized by numerous studies adds to this doubt. I argue that productivity fundamentals still matter and remain the core driver of sustainable growth. However, external factors are crucial for understanding deviations from the trajectory of sustainable growth, i.e. episodes of growth accelerations/decelerations.
Challenges for Understanding Growth in Emerging Markets
As we enter the 4th decade of economic transition in Central and Eastern Europe (CEE), the causes and directions of causality of long-term growth in emerging markets might need to be reconsidered. Some recent studies emphasize that growth trajectories in emerging markets are pretty similar, i.e. average growth rates do not differ too much, while jumps and drops in growth rates are synchronous for the bulk of emerging economies (e.g. Fayad and Perelli, 2014). For instance, a decade ago the level of GDP per capita (in 2011 international $) in Macedonia was roughly 45% of that in the Slovak Republic, which likely reflected the productivity (measured through the Global Competitiveness Index) gap between them. During the last decade, Macedonia has roughly closed this productivity gap. Growth theory would postulate that this should have transformed into faster output growth in Macedonia vs. Slovak Republic closing well-being gap. However, the two countries’ had throughout the decade roughly equal average output growth and the well-being gap today is still the same as it was ten years ago.
Such observations seem to conflict with existing theoretical views. First, this is a challenge to the well-being convergence concept that results from growth theory. Moreover, if we measure growth in terms of the speed of closing the well-being gap with respect to the frontier (the US economy), one may argue even for divergence. For instance, Figure 1 presents a scatter-plot for a sample of emerging markets relating the initial conditions – well-being level in 1995 (GDP per capita relative to one of the US economy) – and the average speed of well-being gap (vs. the US economy) closing throughout 1996-2017 (measured in p.p. of corresponding gap ).
Second, the evidence that productivity gains do not automatically trigger output growth challenges a common view that productivity is the major driver for sustainable growth.
Figure 1.Starting Conditions and Well-Being Gains
Source: Own computations based on data from World Development Indicators database (World Bank).
What are possible explanations for the observed similarity in growth rates of emerging markets?
A study by the IMF (2017) suggests a response: growth in emerging markets is similar and synchronous due to the external environment. This study emphasizes the crucial dependence of medium-term growth in developing countries on the following factors: growth of external demand in trade partners, financial conditions, and trade conditions. Moreover, it states that these factors are dominant in explaining the episodes of growth strengthening/weakening.
Does this explanation change the growth nexus for emerging markets? Can one state, that while external factors are crucial for growth and growth in developing countries is rather homogenous, the productivity gains are not so important anymore?
I would say no. First, for better understanding of growth patterns we must clearly compare the relative importance of productivity gains vs. external factors in affecting the growth schedule. Second, we must separate relatively short-term fluctuations in GDP growth from sustainable growth.
Detecting Relative Importance of Growth Drivers
To answer the question about the relative importance of productivity fundamentals and growth factors, I study a panel of 34 emerging market economies (EBRD sample netted from 3 countries for which the data is not available) for 11 years (2007-2017).
To evaluate the relative importance of productivity and external factors, I use a standard approach of running panel growth regressions with fixed effects. At the same time, I make a number of novelties in the research design.
First, for measures of productivity, I engage a unique database – Global Competitiveness Indicators by World Economic Forum (WEF). Although this database provides an insightful perspective on productivity fundamentals at the country level, it is rather seldom a ‘guest’ in economic research. From this database, I extract a number of individual indicators in order to detect which ones among them that have the strongest growth-enhancing effect. For an alternative specification, I use principal components of 9 individual indicators from this database as proxies for productivity gains.
Second, for external factors, I use an approach similar to the IMF (2017) and calculate variables representing external demand growth, trade conditions, and financial conditions (such as a measure of capital inflows) for each country. Moreover, in respect to external demand growth, I use different competing measures (based on either imports of GDP growth of trade partners) and choose the best one in each individual equation. By doing so, I allow this dimension of the external environment to be represented in each model to the largest possible extent.
Third, I depart from using output growth as the only measure of economic growth and response variable in growth regressions. I argue that for international comparison purposes it is worthwhile to consider also the speed of closing the gap towards the frontier (the US economy). On the one hand, this measure is strongly correlated with the traditional output growth rate. On the other hand, this measure, in a sense, nets out the growth rate of a country from global growth, thus capturing something more unique and peculiar just to individual countries’ gains in well-being. Furthermore, I argue that in the discussion about the factors behind growth, one should distinguish between relatively short and long term growth. Annual growth rates, especially at relatively short time horizon, are too dependent on fluctuations, which may be interpreted in terms of growth rate strengthening/weakening. However, to emphasize the property of growth sustainability, we should get rid of ‘unnecessary noise’. For this purpose, I also introduce a trend growth rate measured in a most simple way as the 5 year moving average (following the discussion in Coibion et al. (2017), show that the bulk of measures of ‘potential’ growth are not good enough to get rid of demand shocks and these measures are pretty close to simple moving average measures).
I apply this definition of trend growth both to ‘standard’ GDP growth rate and to the speed of closing the gap towards frontier. So, finally I have 4 response variables: ‘standard’ growth rate, the speed of closing the gap to frontier, and two corresponding measures of trend growth.
Sustainable Growth Mainly Depends on Productivity
Having short-term (annual) growth rate as response variable (either ‘standard’ or the one in terms of closing the gap) provides results close to those in IMF (2017). It may be interpreted in a way that the external environment is more important than productivity factors. If dividing all regressors into two broad groups of factors – external and productivity – the former is responsible for up to 70% of the growth effect, while the latter for about 30%. Among external environment factors, the most important one is financial conditions. Its relative importance is roughly 50% of the group of external factors’ total.
Among productivity fundamentals, an important contributor to short-term growth is the quality of the macroeconomic environment. According to the methodology of WEF (2017), this indicator encompasses the fiscal stance, savings-investment balance, the external position, inflation path, debt issues, etc.
When refocusing from short-term growth to the growth trend as a response variable, the relative importance of the factors behind growth changes. Productivity fundamentals in this case drive up to 80% of growth effect, while external factors are responsible for the remaining 20%. It is worth noting here that the proportion in favor of productivity factors is higher for the concept of closing the gap to frontier rather than for ‘standard’ trend growth rate. This evidence may be interpreted as additional justification for treating this measure of growth as ‘good’ at reflecting individual properties of a country in a global landscape.
Furthermore, the role of individual variables also changes. Among external factors, the most important role in driving sustainable growth belongs to trade conditions and external demand growth, while the role of financial conditions is either miserable or insignificant at most. Among productivity factors as drivers of trend growth, the quality of the macroeconomic environment seems to play a special role, as well as the efficiency of the goods market and the financial system.
The evidence showing rather similar and synchronous growth in emerging markets and recent evidence on the crucial importance of external factors for emerging markets should not lead us to incorrectly believe that productivity fundamentals do not matter anymore. Productivity fundamentals are still the core driver of sustainable growth. At the same time, we should keep in mind the important role of the external environment for emerging markets. However, changes in the external environment are more likely to generate relatively short-term growth rate fluctuations, while having a modest impact on the sustainable growth trajectory. Hence, a country aiming to secure sustainable growth should still first of all think about productivity fundamentals.
- Coibion, O., Gorodnichenko, Y, Ulate, M. (2017). The Cyclical Sensitivity in Estimates of Potential Output, National Bureau of Economic Research, Working Paper No. 23580.
- EBRD (2017). Transition Report 2017-2018, European Bank for Reconstruction and Development, London, UK.
- Fayad, G., and Perelli, R. (2014). Growth Surprises and Synchronized Slowdown in Emerging Markets—An Empirical Investigation, IMF Working Paper, WP/14/173.
- IMF (2017). Roads Less Traveled: Growth in Emerging Markets and Developing Economies in a Complicated External Environment, in IMF World Economic Outlook, April, 2017, pp. 65-120.
- World Economic Forum (2017). The Global Competitiveness Report 2017-2018, Geneva: World Economic Forum.
Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.
This brief summarizes the results of research on the political costs of large-scale economic crises. In a large historic sample of countries, we study the impact of different types of crises, such as sovereign and domestic defaults, banking crises and economic recessions, on political turnover of top politicians: heads of the state and central bank governors. According to the findings, only default on domestic debt increases the probability of politicians’ turnover but not the default on external debt. As argued, this is due to the fact that the latter is not directly felt by the voters. In addition, we find that although currency crises increase chances of head of central bank turnover, it does not affect tenures of heads of state. Presumably, this is the case since currency crises are in the eyes of the public the responsibility of CB governors. These findings are relevant for both developed and transition economies, but are especially important for the latter as political turmoil and economic recessions are more prevalent in developing nations.
Overview and Key Findings
Large-scale economic crises are associated not only with the economic downturns, but also with political turnover. When the national economy is in a critical state, a default declaration often turns the economy back to growth as it is typically viewed as an act of acknowledging a problem and showing readiness for changes. However, politicians responsible for the economy and leaders of the states are often reluctant to declare default and try to postpone it, which worsens the situation. One of the reasons behind such unwillingness to act is a fear of a political turnover following the open acknowledgement of a problem.
This brief summarizes the findings Lvovskiy and Shakhnov (2018). We investigate the statistical evidence of political costs related to different types of economic crises.
We find that the effects of a crisis depend on the crisis type and on whether it was in the area of responsibility of a given politician. For example, external sovereign defaults have no effect on political turnover, which we interpret as external sovereign default having a small impact on the general public. On the contrary, domestic sovereign defaults have a large impact on the country population and often lead to the replacement of the top executive. In turn, banking crises are followed by the downfall of the government at the level of chief executive as well as the governor of the central bank.
While there is large literature on career concerns of politicians and political turnover, the majority of papers either focus on the regular changes through elections in democratic regimes (Treisman, 2015) or study a particular non-democratic country, like China (Li and Zhou, 2005). However, throughout history, crises have often happened in transition, non-democratic or not fully democratic countries. Furthermore, even in democratic countries many changes of government have been irregular. Since a delay in default declaration usually harms economies it is important to understand the mechanisms behind it in different institutional settings. Our paper contributes to this understanding by analyzing the impact of economic crises on political survival in a wide set of countries and regimes. Better understanding of the political costs that the top executives face while making such decisions is crucial for the prediction of these decisions as well as for international default negotiations and consultations.
Below we describe our finding in some more detail.
Statistical Analysis and Results
Our analysis consists of two main parts. We start with the political turnover for heads of state, who are in charge of the performance of the whole economy, which we measure by the GDP growth. Then, we look at central bank (CB) governors, who are in charge of the monetary policy, price stability, stability of the financial sector and banking supervision.
Table 1. Head of state changes
Table 1 presents the estimated linear probability regression models for the head of state turnover. As expected, elections have a strong impact on the probability of the turnover of the head of state. Further, as Column 1 in Table 1 shows default on external debt has no significant impact on the head of state tenure while default on domestic debt increases the yearly chances of being displaced by 34 %. This supports the idea that voters care more about their own savings than about the general situation with the state’s budget. When we look at the effect of past crises (the predictor variable in this case is whether a crisis took place last year), Column 2 coefficients for both external and domestic defaults appear to no longer be statistically significant. Instead, banking crises become significant. This situation could be due to the fact that one of the common consequences of domestic defaults is an ongoing distortion of savings which often leads to deposit runoffs, so the effect of the previous year’s domestic default now acts through a banking crisis.
Table 2. Central bank governor changes
Table 2 presents similar results but this time the left hand side variable is CB governor turnover. Similarly to the case with the head of state turnover, only default on domestic debt has a significant effect on the CB’s governor tenure and not the one on external debt. The main differences with Table 1 are that elections do not statistically predict turnover of CB heads while currency crises do. The former result is expected since in most countries there are no direct elections of central bank governors and central banks often have some degree of independence from the government. The latter result, that currency crises have a significant impact on CB governors’ tenures, implies that since currency control is one of the roles of a CB, its head is held accountable for currency crises and not the head of a state.
We examine the political cost of different types of economic crises, and find non-uniform effects of different types of crises on the political survival of various key officials. Domestic defaults, and recent banking crises are shown to be costly both for heads of states and central bank governors, while currency crises only have an impact on the political survival of the latter.
Interestingly and importantly, we find no evidence of the impact of (external) sovereign default on political turnover of the head of state or central bank governors. In other words, contrary to Yeyati and Panizza’s (2011) suggestion, it seems that there is no immediate political cost at the top associated with (external) sovereign default. One possible explanation is that the public does not punish a politician for defaults because by defaulting, the politician makes the optimal decision. In a modern world, many developing nations experience rapid growth of their sovereign debt. The presented evidence brings partial optimism that even if economic mistakes have already been made, top politicians would understand that acknowledging a problem and making steps toward its solution may not always be as costly for them as has previously been thought.
- Li, Hongbin; Li-An Zhou, 2005. “Political turnover and economic performance: the incentive role of personnel control in China,” Journal of Public Economics, 89 (9), 1743 – 1762.
- Lvovskiy, Lev; Shakhnov, Kirill, “Political Responsibility for Different Crises”, BEROC working paper #50, 2018
- Treisman, Daniel “Income, Democracy, and Leader Turnover”, American Journal of Political Science, 2015, 59 (4), 927–942.
- Yeyati, Eduardo Levy and Ugo Panizza, “The elusive costs of sovereign defaults,” Journal of Development Economics, January 2011, 94 (1), 95–105.
Author: Tom Coupé, KSE.
Ukraine needs reforms badly. However, there is a huge difference in how the government, the expert community, and the general public understand reforms. According to a recent survey conducted by a prominent Ukrainian newspaper, people expect that reforms should, in the first place, improve their personal wellbeing. However, research findings beware that in the short run structural changes in the country can worsen economic performance and increase inequality. To reduce the pain of unmet expectations and popular discontent, the government should openly communicate any difficulties to come, and wisely mix the most painfull measures, like the increase of tariffs for the use of public infrastructure, with empowering changes that give citizens a sence of progress, like actions that strengthen democracy and help SMEs to flourish.
Over the last couple of years, the growth rate of potential Belarus’ GDP declined. The government intends to revive economic growth by the policy of ‘modernization’, in practice pinned down to a drastic increase in the volume of capital investment, including by the means of directed lending. As the pre-crisis macroeconomic imbalances are at least partially cured, the government seems to be eager to apply a familiar policy tool. However, the empirical analysis of the effects of directed lending on total factor productivity and economic growth casts serious doubts on the efficiency of this policy tool.
Over the last couple of years, the growth rate of potential Belarus’ GDP declined. This conclusion is robust as suggested by the application of competing methodologies to assess potential GDP. For instance, the statistical filters, including the HP-filter, the Kalman filter, and the production function approach, produce different levels of potential growth, but generate similar growth rate dynamics, particularly the downward trend. From this perspective, the tendency for high and sustainable GDP growth in Belarus is increasingly compromised.
Economic authorities seem to be aware of that fact. For instance, the Ministry of Economy stresses the need to create a new, ‘highly productive’ sector in the national economy as the new engine of growth. An ambitious plan involves expanding the size of this sector to contribute to about half of the GDP growth rate, aimed at 12 per cent per annum by 2015. The creation of this ‘highly productive sector’ falls into recent policy initiative, called ‘modernization’. Under this banner, the government plans to renovate the capital stocks (primarily machinery, equipment, and transport vehicles) of a large number of state-owned enterprises. In a nutshell, this strategy may be seen as a way to facilitate technical progress embodied in capital.
What is necessary, according to the government, is to make a spurt in capital investments, often on a case-by-case basis. The government has a pool of enterprises to be modernized. The majority of them are unable to modernize themselves – i.e. radically increase capital investments – due to the lack of internal funds and poor access to external finance. Accordingly, directed lending is considered to be a useful policy instrument of modernization. In 2013, the Development Bank plans to considerably increase its credit portfolio (by about USD 0.5 billion) by financing projects at subsidized interest rates under the ‘modernization’ program. Recently, the government compiled a list of 67 agricultural enterprises liable to have an access to cheap loans for modernization purposes from the Development Bank. In addition, state-owned banks will continue the provision of policy loans that can be considered as directed ones.
With directed loans, we mean those loans that are typically granted to selected borrowers at interest rates lower than the market interest rates. In Belarus, directed lending has been an important policy tool over the last decade. Selective credit programs have been applied to prevent underinvestment and to stimulate output growth.
According to the estimations of Fitch Ratings (2010), almost a half of the outstanding loans in the Belarusian economy by the end of 2009, were directed ones. The IMF provided a slightly smaller, but still substantial figure of 46.2 percent (IMF, 2010). According to our own calculations, by 2011, the volume of directed loans amounted to about 40 percent of the total volume of outstanding loans. These loans have been made abundant in agriculture and housing construction sectors and, to a lesser extent, in manufacturing. This massive presence of selective credit in the national economy can be seen as a large factor contributing to the currency crisis of March 2011.
Accordingly, after the crisis, and following the necessity to ‘clear up’ the assets of the national banking system, the share of directed lending was reduced. We estimate that in 2012, the ratio of directed loans in total loans dropped to roughly 30 percent. However, the recent rhetoric of the development of ‘highly productive’ sectors and modernization is indicative of the intention to find new life for this old cloth. Directed lending is expected to revitalize enfeebling growth. In 2012, real GDP growth amounted to 1.5 percent against the background of the initial government plan of 8.5 percent.
Under selective credit programs, banks have been partially deprived of their autonomy to make decisions over the provision of credit. Thus, banks’ intermediation role has been circumscribed by the authorities. In theory, directed loans may spur capital accumulation as beneficiaries of these loans have access to cheap loans and thus invest and – arguably – produce more. In Belarus, there has also been an additional incentive, i.e. the necessity to substitute depreciating and outdated capital stock, inherited from the Soviet past. At the same time, political interference into the process of credit provision suggests that loans may be allocated to lower-yielding projects, and thus dampen growth rates of factor productivity and GDP (Fry, 1995). In addition, non-favored companies – typically from the private sector – face higher interest rates as their state-owned counterparts receive substantial discounts for their use of capital.
So far, these soft budget constraints in the financial system have allowed favored companies to receive loans up to three times cheaper, if judged by the level of real effective interest rates. Although private companies tend to be more efficient than state-owned enterprises in terms of factor returns and profitability, higher interest rates may reduce the volume of outstanding market loans. Furthermore, increases in the volume of cheap residential loans, which do not contribute directly to enhancement of productive capacity of the economy, may dampen the returns on investment further.
Governments have traditionally relied on selective credit programs by stressing positive externalities and spillovers for the economy as a whole (DeLong and Summers, 1991). Commercial banks care about private returns, while governments seek to maximize social returns by financing firms, which are capable of generating positive externalities. Unfettered operations of credit allocation mechanisms minimize allocation inefficiency and induce banks to minimize the costs of financial intermediation, thereby making credit more accessible.
How do these competing forces meet in Belarus and what are the effects of their joint working? In answering those questions, we have conducted an empirical analysis of the effects of directed lending on total factor productivity dynamics. The latter is considered to be a good proxy to observe the impact of selective credit programs on the efficiency of actor use.
The results of our econometric analysis show that over the period concerned, 2000–2012, the expansion of directed lending in Belarus has negatively affected total factor productivity dynamics and, subsequently, negatively contributed to the rates of GDP growth. A positive impact on growth, stemming from additional capital accumulation might nevertheless occur, but with a substantial lag. This likely positive impact is associated with the ability of banks to increase the volume of market loans alongside with the rising volume of directed loans. The option has been made possible only due to massive liquidity injections by the government and mainly the National Bank of Belarus. However, such injections are problematic to maintain over the medium to the long run as they have severe inflationary repercussions for the economy.
The effects of individual components of directed lending are mainly the same. In particular, loans for residential construction, provided to households in need, negatively affect total factor productivity. Moreover, it is through housing loans the adverse effects of directed lending upon factor productivity are mainly realized. The interest rate spread – between preferential interest rate and market interest rate – amplifies these negative relationships. Lower preferential rates result in larger losses in total factor productivity. Loans to agricultural firms have similar impact, although it has to be emphasized that the overall impact on total factor productivity approaches zero (not negative, as in the case of housing loans).
We also find that for Belarus, an increase in the total volume of directed loans leads to an increase in the volume of market loans. Both the National Bank and, to a lesser extent, the government, strive to minimize risks in the national banking system, which provide loans with smaller returns and/or non-performing policy loans. Similar challenges have been observed in China, where the Central Bank has been forced to recapitalize domestic banks to support economic growth after the global financial crisis of 2008. In 2007–2008, Chinese growth of 8–10 percent was driven by new lending averaging 30–40 percent of GDP, of which up to a quarter of the loans might have been non-performing, amounting to losses of 6–10 percent of GDP (Das, 2012).
In Belarus, the recapitalization policy, apart from its inflationary consequences, has other important effects. In particular, it prevents a dangerous trade-off between directed loans and market loans to resurface, whereby the former crowds out the latter as banks are unable to expand their portfolios due to the liquidity constraints.
Therefore, unless the expansion of directed loans would be checked, adverse effects of selective credit programs on productivity and growth would not evaporate, with negative consequences for the whole economy. Regarding policy recommendations, we claim that there is a need to fundamentally revise directed lending policies or to even minimize it to the extremes by allowing standard market mechanism for credit allocation to prevail in the national economy. Furthermore, we argue that directed lending, even after some cosmetic changes in the system design made in 2012, is not an efficient tool for economic growth promotion.
Tentative results of growth accounting made at the level of selected important industries suggest that the downward growth dynamics is associated with weak total factor productivity growth, i.e. disembodied technical progress. Improvement of total factor productivity seems to have the biggest potential for revival of economic growth. Therefore, the use of directed lending, as a policy instrument that hampers total factor productivity dynamics, may undermine prospects for long-term economic growth in Belarus.
- Das, S. (2012). “All Feasts Must Come to an End– China’s Economic Outlook”, Euro Intelligence, 11 March, viewed 12 April 2012.
- DeLong, J.B. and L.H. Summers, (1991). “Equipment Investment and Economic Growth”, Quarterly Journal of Economics 106, 2, pp. 445–502.
- Fitch Ratings, (2010). “Directed Lending: On the Up or on the Way Out?”, Belarusian Banking Sector, May.
- Fry, M.J. (1995). Money, Interest, and Banking in Economic Development (John Hopkins University Press, Baltimore and London).
- IMF (2010), “Republic of Belarus: Fourth Review under the Stand-By Arrangement”, IMF Country Report 10/89, viewed 15 July 2012.