Japan 3Q18 GDP - blame it on the weather

Our initial stab at creating a 3Q18 GDP forecast for Japan some weeks ago used the data then available, predominantly real household spending data, and core machine orders, together with some housing and other construction data.

While the housing numbers were on the soft side again, the bulk of the rest showed an overall improvement from 2Q18, and this despite what was a very sharp downwards plunge in September, as Typhoon Jebi, the strongest typhoon to hit Japan in 25 years, wreaked havoc on the west of the country, with flooding and damage to airports roads and bridges on top of the human casualties.

Taking a neutral view on trade, the data suggested another decent quarter for growth. This now appears to be quite the wrong conclusion to have drawn.

Since then, real seasonally adjusted import and export data has been released, and this shows a large fall in exports on the quarter, together with a big rise in imports - some of this is likely to show up as higher inventories somewhere in the GDP accounts, but nonetheless, we are likely to see a substantial hit to GDP coming from the trade side.

The consensus forecast for Japan is pretty tight, with the QoQ figures clustered mainly in the -0.2 to -0.4 range, for a median annualized growth rate of -1.0% (we forecast -0.1%QoQ and -0.4% QoQ annualized). This would be quite a climb down from the +3.0% annualized growth recorded in 2Q18, if that remains unrevised.

Our own forecasts remain somewhat higher than the consensus. The positive results from real household spending and machine orders don't map all that well onto the GDP-based data, so there is far more room for surprise than the consensus cluster suggests. Moreover, typhoon-affected trade data, we believe, is more likely to see partial inventory offsets that will soften the blow from the net export figures - which themselves require a bit of interpretation from the higher frequency data.

The final message to take away from 3Q18 GDP is that, no matter how bad, any dip is likely to be rapidly reversed in 4Q18, so the impact on full-year growth is likely to end up being insignificant. Indeed, with clear up activity and rebuilding, replacement of damaged goods and infrastructure, we may end up nudging up the 2019 GDP forecasts.

None of this, however, should have any material impact on the Bank of Japan, and their policy stance, and therefore, for the JPY.