Russia: September activity interesting but uninspiring

Even though retail trade growth decelerated from 0.8% year-on-year in August to 0.7% YoY in September, instead of accelerating to 1.0% YoY expected by us and consensus, we do not take it as a sign of strong deterioration in the consumer trend.

• The slowdown was driven by the food segment, while consumption of durable goods posted some acceleration
• More modest consumption seems to reflect a higher preference for savings (for the second month in a row), as retail deposit growth (adjusted for revaluation of US$96 bn worth of FX deposits) accelerated to 9.3% YoY in September, a 27-month high, while retail loan growth continued to gradually slide to a 12-month low of 20.7% YoY
• The strength of retail trade as a proxy for overall consumption growth has declined due to the development of internet trade and services. To remind, there was acceleration of consumption growth from 1.2% YoY in 1Q19 to 2.1% YoY in 2Q19 despite a 0.3 percentage point deceleration of retail trade.

In the meantime, we also doubt that accumulation of saving will translate into a significantly stronger consumption trend anytime soon. First, starting 1 October, banks are facing higher risk weighting on unsecured retail loans to clients whose interest payments account for 50%+ of their income, which should affect the overall retail lending trend and limit consumption growth by the lower income households.

Second, the sustainability of improvement in the overall income trend is questionable. While Rosstat (the official statistics service) reported 3.0% YoY growth of real disposable income in 3Q19 following a 0.1% YoY contraction in 2Q19, the source of this improvement remains unclear. The statement by Rosstat suggests that salary growth was the key driver, however it has yet to provide a data set to support this claim. Rosstat has enforced a one-month gap in reporting official salary growth, the most transparent indicator, meaning a delay in the release of September data till November. Since the publicly available data suggests there was no improvement in real salary growth in July-August (2.4-3.0% YoY) vs. 2Q19 (2.6% YoY), a spike in the overall real disposable income growth indicated by Rosstat would require any (or some combination) of the following:

1. A massive acceleration of nominal salary growth from 7.2% YoY in 8M19 to 13-17% YoY in September, ie, to rates not seen since 2013. If this were the case, it would likely reflect some one-off payments in one of the sectors or would suggest a change in methodology. Rosstat's official representative indicated to us by phone that the 9M19 nominal salary growth of 8.1% YoY indicated in the press-release is not comparable with the 7.2% YoY reported for 8M19, as the former is a measure of another indicator, that incorporates social benefits to workers, unlike the latter. A set of comparable data was promised to be released later this month.
2. A massive increase in the number of officially employed population receiving the indicated salary. The Rosstat data on employment does indeed show a 0.7 million increase in the number of employees in 3Q19 vs. 2Q19, however this increase is in line with seasonality (summer immigration). In fact the number of employed in 3Q19 was 0.8 million lower than in 3Q18 on deteriorating demographic trends, which goes against acceleration of annual growth rate in the labour income.
3. A massive pick-up in other sources of income. Those include unofficial salaries, pensions and other public benefits, or interest income. Together they account for around 44% of the household income and in theory could result in an improvement in the overall income growth. However, the data provided by Rosstat does not confirm this theory, as the reported nominal growth of income from those sources at 5.8-7.3% YoY in 3Q19 underperforms the overall nominal income growth of 7.8% YoY. This fact makes sense given the depressed mood in SMEs, lack of aggressive indexation of public sector payouts following the end of electoral cycle in 2018, and the declining interest rates on deposits and other financial assets held by the population.

Pending clarification on the official income data by Rosstat, we advise to follow Bank of Russia's monthly statistics on retail lending and retail deposits as more transparent and consistent indicators of household behaviour.

On a more dull note, the producer trend failed to bring surprises. A slight pick-up in construction growth from 0.3% YoY in August to 0.8% YoY in September is indeed an improvement vs. flat-lining 8M19, however, as is the case with industrial production we commented on earlier, it reflects a favourable calendar factor (extra workday in September 2019 vs. September 2018) rather than anything else.

The lack of improvement in corporate activity is highlighted by a strong deceleration in corporate loan growth (adjusted for FX revaluation effect) to a 15-month low of 3% YoY in September. The recent corporate newsflow suggests a continued accumulation of profits as local and international financial assets, and redistribution of a larger share of profits as dividends, rather than capex.

Overall, we still believe the corporate sector will find support from the budget spending, which should accelerate from to 6% YoY in 9M19 to 17% YoY in October-December (and from 4% YoY to 13% YoY in the segment of targeted support to the economy), but should the main bulk of it be delayed until the very end of the year, the favourable effect on producers may be postponed till next year.

Given the above-mentioned considerations, we doubt that Russia will be able to maintain the 1.9% YoY preliminary GDP growth estimated by the government for 3Q19. The sharp acceleration vs. 0.9% YoY in 2Q19 is itself a bit of a question mark given the lack of improvement in retail trade and industrial output dynamic. According to our estimates, out of an overall 1 percentage point acceleration, 0.2 percentage point reflects a pick-up in agricultural output on favourable weather and harvest conditions, however that support factor should be considered temporary. The rest of the growth structure remains non-transparent and requires further clarification.

We would not exclude that the initial GDP estimate for 3Q19 might be revised downward by Rosstat, and expect sustainable acceleration of GDP growth starting closer to 2021-22, when the infrastructural spending on the National Projects gains traction.