Taking the wind out of sales
17 April 2018
Japanese consumption has had a stuttering start to the year, with consumer confidence and retail sales correcting in the first few months (see Chart 8). So far, we have been comfortable with the idea that the weakness is temporary, reflecting bad weather. However, how confident are we of a rebound and what does this suggest for the outlook for growth in Q1? In terms of the composition of retail sales, the most significant negative contribution came from apparel and autos. The former slumped 3.6% in the January-February period versus the previous quarter, while the later was down 6.3%. For the apparel sector to return to its Q4 average it would need to jump 9.4% m/m, while the auto sector reading would need to jump more than 20% m/m. The partial data for March point to a far more cautious tone, with department store sales data modestly weaker while new auto sales for the month dropped 1.0%. So should we be downgrading our expectations for consumption in Q1?
It is worth reminding ourselves, retail sales often diverge meaningfully from the national accounts data as they include sales to overseas travellers. A more comprehensive guide to expenditures as well as incomes comes from the Family Income and Expenditure Survey. The latest household survey saw real consumer spending (ex-housing) fall 2.2% m/m in February. However, a 2.6% m/m jump in January means that the January-February average was up 1.6% on the Q4 average. So, are worries about a consumption slowdown overblown? Two key factors were driving the outlook for consumption to remain robust in 2018. Firstly, workers’ income had stabilised at a reasonably high level. Secondly, households’ propensity to consume appeared to have bottomed. Both of these factors have been challenged in recent months (see Chart 9). Before we take the red pen to our forecasts, it is worth noting that changes in data collation methodology in the Household Survey introduced in January, including the addition of a full-time employee/worker households sub-category and the listing of median values as well as averages, has made it difficult to make historical comparisons. This leads us to alternative partial indicators of underlying spending trends.
One of our preferred gauges of household consumption is the Cabinet Office’s Synthetic Consumption Index. The latest reading pointed to a modest improvement in February, up 0.6%. That leaves the January-February average broadly flat against Q4. Unfortunately, the index gives little additional detail at a sub-component level, so it is difficult to pinpoint any source of the strength. The Bank of Japan’s Consumption Activity Index sheds a little more light. While real consumption is broadly flat from Q4, on an aggregate basis the Services Index in January-February was 1.4% higher than the previous quarter. So, is services spending set to rescue household activity in Q1? Typically, we would expect stronger services spending to show up in a better reading for the territory activity index. That this has not been the case means that we now expect private consumption to be flat on the quarter, meaning a modest downward revision from our 0.1% q/q pencilled in previously. Combined with recent weakness in industrial output our 2018 forecast comes down from 1.2% to 1.0%. This remains a fairly modest adjustment and our forecast is still for activity in Japan to remain comfortably above potential.