- Russians have been enjoying a steady average debt load decline for almost a year and a half. From a peak level of mid-2014 when servicing and potential loan repayment accounted for 9% of population’s cash incomes this indicator dropped to current 8.5%, although average credit quality of individuals is low and payment discipline with regard to old loans is still falling. A resumption of population’s debt load growth will result in an excessive risk of financial stability loss.
- The expected decline in interest rates will provide for an increase in lending with no further debt load accumulation. Slashing interest rates by 4 pps should entail more than 9.1% of average annual growth of private individuals’ loan portfolios in 2016-2017.
- While real incomes stagnate, demand on consumer markets will depend on lending in 2016-2017. Effective demand growth rates can reach 0.4% per annum in real terms. Should the current level of population’s debt load stay flat, the actual demand growth on the housing market will hardly exceed 8.1% and the automotive market may see even a lower figure of 0.4%, while consumer markets, except the automotive one, may show no growth at all.
Russians see their debt load steadily decline for the first time in six years upon reaching the developed countries average
See ACRA’s report “Consumption: absence of growth to be offset by quality change” published
June 6, 2016
A rapid growth of lending to individuals in 2011-2014 supported economic growth in Russia, and starting 2013 it stimulated consumption despite absence of real income growth. Although by 2016 the households’ debt equaled just 13-14% of GDP (i. e. it was much lower than in an overwhelming majority of countries with a developed financial system), their debt load is relatively high. By early 2014, Russia’s debt service ratio (DSR) has reached the developed countries average of 9%. Since then, the size of population’s debt burden, coupled with tighter regulation of the sector, has had a retarding effect on lending (Figure 1).
Figure 1. Debt servicing to disposable income ratio in Russia compared to individuals’ debt load in other countries
Debt service ratio (DSR) is the proportion of disposable income required to service interest payments and repay the short-term part of debt. Data for 18 developed countries is a BIS estimate. Data for BRIC countries is an ACRA estimate based on official statistics and a methodology comparable to the one used by BIS. Hereinafter, the term “debt load” shall be used as a synonym of DSR.
With no income growth in place, building up debt further may be fraught with a dangerous decline in credit quality of population. Over the last three years, the share of loans with payments overdue for more than three months has doubled surging from 5% to 10.8% (Figure 2). Moreover, this indicator has moderated its climb but not halted it after the loan portfolio stopped growing, which implies a continuously deteriorating payment discipline regarding old loans.
The main mechanisms limiting the debt load level of the population are deemed to be risk management policies in the banking system and Bank of Russia’s financial stability policy. Studies based on international experience show that in two cases out of three a 4-6 pps increase in debt load over the long-term average precedes a financial crisis (See the article “Do debt service costs affect macroeconomic and financial stability?” by M. Drehmann and M. Juselius, 2012). In Russia, a situation close to this alarming scenario still persists, so in the coming years the regulator may try to prevent a new growth in individuals’ debt load. Some serious steps have already been taken during the credit boom, i.e. starting 2012: capital adequacy standards were complemented with a requirement to weigh various asset types depending on their risk, while bank consumer products obtained regulatory limits on the total cost of credit. Additional measures are also under development. On June 21, the Bank of Russia published a draft document, which implies raising risk ratios for unsecured consumer loans.
Aggregate debt load of individuals and the non-financial sector in Russia climbed 3.5 pps relative the long-term average in 2013-2015.
Figure 2. High debt load amid non-growing incomes may entail a drop in credit quality
In view of possible further creditability decline of the population, the latter’s debt load is likely stay flat at current levels in the next 2-3 years if we assume a worst-case scenario, but a more likely paradigm favors debt load’s gradual decline through a steady “dilution” of the share of poorly serviced loans disbursed at the peak of the economic downturn.
Lower interest rates will boost lending to individuals, while keeping their
debt load flat
A gradual slowdown of observed and, more importantly, expected inflation will result in lower medium- and long-term lending rates in the next two years. This will enable individuals to build up their bank debt while keeping debt burden flat. To date, some 45-46% of debt servicing is in fact almost entirely represented by interest payments, while the rest falls to the share of repayment or refinancing of short-term debt. A decline of the first of these components through lower interest rates should make building up loan portfolios possible.
The assumptions used are deemed to be realistic if inflation reaches the Bank of Russia’s target of 4% by 2017.
Assuming that debt load of the population stays at current levels (an “extreme” scenario; in fact, it is likely to decline, as discussed above) while interest rates decline, we can assume the extent of bank loan portfolio growth in the next two years. If nominal incomes of the population climb 5% per year and lending rates lose 4 pps over the period of December 2017 to December 2015, the portfolio may grow 9.1% each year, which corresponds to a two-year total of RUB 2 trillion or a 3% average annual growth in real terms. That said, the contribution interest payments make to population’s debt load may shed up to 2 pps each year (Figure 3).
Figure 3. With interest rates falling, interest payments will have a weaker impact on debt load
Terminal growth rates of the bank portfolio based on various assumptions of private incomes and lending rates (with debt load remaining flat) are shown in Tables 1-3.
Table 1. Upper estimate of average annual growth of individuals’ debt to banks in 2016-2017, %
Table 2. Upper estimate of individuals’ debt to banks growth in 2016-2017, RUB bln
Table 3. Upper estimate of average annual growth of individuals’ debt to banks in 2016-2017 in real terms (CPI adjusted), %
Table 4. Upper estimate vs base-case scenario
Real growth in effective demand through lending is possible on the housing market and is unlikely in other segments
See an ACRA report “The Russian economy: absence of usual conditions for growth offset by impetus for structural change” published March 21, 2016.
With real incomes stagnating or even declining, lending will be the key factor determining demand for consumer goods, cars and real estate. Using the calculated above upper limit of population’s loan portfolio growth and the current cash income growth estimate, we get the maximum assessment of effective demand growth on selected markets in 2016-2017 (Table 5).
If the current loan purpose structure remains and households use additional current income to finance expenses like they did last year (Figure 4), the real growth of the car market will likely be capped at 0.4% per year and the real estate market will hardly climb higher that 8.1% per year, while consumer markets (except the car market), may show no growth at all.
Figure 4. Lending growth and household spending by purpose in 2010-2014
Table 5. Effective demand maximum growth estimates for key consumer markets
Calculation method and notes
1. DSR (debt service ratio) for the population has been calculated based on the methodology (See the article “How Much Income Is Used for Debt Payments? A New Database for Debt Service Ratios” by M. Drehmann, A. Illes, M. Juselius and M. Santos, 2015) similar to the one used by the Bank for International Settlements (BIS):
where stands for interest payments per year, – for planned principal repayment per year, in line with the repayment schedule (refinancing and term revisions are not accounted for), and – for disposable income per year in current prices.
In case of data absence regarding the actual repayment schedule and term structure of individuals’ debt, a simplified assumption of an annuity payment form and a corresponding approximate formula were used:
where stands for the average rate of disbursed loans for three preceding years, – for individuals’ bank loan portfolio as of the beginning of period t, – for an average term of current loan agreements as of the beginning of period t and – for disposable income per year in current prices.
For Russia, this approximation produced a less than 0.2 pps inaccuracy for quarterly data within the 2008-2014 period, so the calculation method using the formula (1) has been recognized correct and commensurable with calculations made for other countries using the formula (2).
In case of Brazil, calculations by the Central Bank of Brazil based on quarterly data were used.
For China and India, the calculations were based on annual banking statistics provided by national central banks and on annual household income data obtained from national statistical agencies, coupled with the annuity payment form assumption and, consequently, with the formula (2) used.
2. The June 2015 study by the Bank of Russia (See the article “Debt Load Indicators” by S. Donets and A. Ponomarenko, 2015) featured similar DSR calculations, but with a different income base of the population used – the Central Bank based its research on data on wages taken from the national accounts section dedicated to GDP broken down by revenue sources. This explains the difference in the absolute level assessment of the individuals’ debt load.
3. To calculate the upper estimate of individuals’ credit portfolio growth in Russia the formula (1) was used. The indicator was fixed at the 4Q15 level, while the was modeled as a product of the average lending rate for the last three years and the loan portfolio volume . The variable was modeled as a product of the short-term debt share and the loan portfolio volume . Both models are considered suitable for forecasting, as they proved to be rather accurate in explaining the actual dynamics of interest and principal repayments within the 2007-2015 period. The share of short-term debt has been fixed for forecasting purposes at the level of 4Q15.