Better-than-expected macroeconomic data overshadowed by NBP currency intervention

Retail sales in November above expectations

In accordance with GUS data that have been released today, retail sales in enterprises employing more than 9 people decreased in current prices by 5.3% YoY in November vs. a 2.1% decrease in October, running above the market consensus (-7.4%) and our forecast (-9.4%). The sales dynamics in constant prices dropped to -5.3% YoY in November vs. -2.3% in October. Seasonally adjusted retail sales in constant prices were lower by 0.6% compared to October 2020. The decrease in retail sales in November and the deepening of their decline in annual terms resulted from the restrictions imposed in November to contain the number of new COVID-19 cases, including in particular the closing of shops in shopping malls. This was reflected by sharp declines in the sales dynamics in constant prices in the categories such as: “textiles, clothing and footwear” (-21.9% YoY vs. -9.7% in October) and “furniture, radio-video and household equipment” (-0.6% YoY vs. 11.9%). The decrease in retail sales was limited by a sharp increase in the share of online sales in total sales (up to 11.4% vs. 7.3% in October), which stood at a level close to the one recorded in April 2020 (11.7%), namely in the period of tough restrictions and low mobility of households. We expect that the opening of shops in shopping malls in December 2020 (pre-Christmas period) will contribute towards a significant increase in retail sales but their real annual dynamics will be close to zero.

Still low activity in construction

According to GUS data, the construction-assembly production decreased by 4.9% YoY in November vs. a 5.9% decline in October, running above our forecast (-8.5%) and the market consensus (-5.2%). The main reason behind the increase in the construction-assembly production between October and November was a favourable difference in the number of working days (in October 2020 this number was 1 day lower from the year before while in November it was 1 day higher than in 2019). The last year’s low base effect had an opposite impact. Seasonally-adjusted construction-assembly production decreased by 0.7% between October and November 2020. A decrease in the construction-assembly production dynamics in November has been recorded in the categories “construction of civil engineering facilities” (-4.5% YoY vs. -4.0% YoY in October) and “construction of buildings” (-12.0% vs. -11.3%), while the category “specialized construction activities” has recorded an increase in production up to 4.7% YoY vs. a 1.8% decrease in October. The data on the construction-assembly production and its structure point to a continuously low activity in construction, mainly due to a sharp decline in corporate investments. We expect low activity in construction to continue until H2 2021 when recovery in corporate investments and faster growth of public investments will contribute to a marked increase in construction-assembly production. Our scenario is supported by a continuingly low, compared to the pre-pandemic level, GUS index illustrating the portfolio of domestic orders in construction companies.

Better-than-expected macroeconomic data overshadowed by NBP currency intervention

The data on the November retail sales and assembly-construction production are yet another – after the data on corporate employment and industrial production released last week (see MACROpulse of 17/12/2020 and 18/12/2020) – signal of a significant upside risk to our forecast of GDP in Q4 2020 (-4.2% YoY). We believe that the new restrictions imposed by the government between 28/12/2020 and 17/1/2021 (closed shops in shopping malls, closed ski slopes, and strong restrictions in hotel activity) as well as travel restrictions on New Year’s Eve and New Year Day are neutral for our forecast of GDP dynamics in Q4 2020 and Q1 2021 (-1.0% YoY). Our macroeconomic scenario for 2021 assumed a “crawling lockdown” in Q1 2021 despite the start of vaccinations against COVID-19 (see MACROmap of 7/12/2020).

A series of better-than-expected data on economic activity in November is conducive to a strengthening of PLN but it is unlikely in the short run due to an increase in global risk aversion (mainly the effect of a very infectious strain of coronavirus being detected in the UK) and the surprising currency intervention carried out by the NBP on Friday with a view to weakening PLN. Due to the aggregate impact of these factors, the EURPLN exchange rate has increased from 4.43 to 4.51. So far the NBP has not released a statement on the intervention but the remarks by MPC members J. Żyżyński and E. Łon after the NBP intervention indicate that its aim was to stimulate exports, with J. Żyżyński pointing to 4.50 as the appropriate level for EURPLN. This justification is surprising in the light of the good results of Polish exports recorded in recent months (see MACROpulse of 18/12/2020). It is also worth emphasizing that, in the light of the studies on the monetary policy transmission, the depreciation of PLN would have to be at a much larger scale from the one recorded to date and would have to be of a lasting nature to substantially increase Polish exports. In our view, the NBP intervention shows that the most likely scenario now is EURPLN amounting to 4.51 at the end of 2020. We also see an upside risk to our forecast of EURPLN in H1 2020 (see MACROmap of 14/12/2020).