The GDP beauty pagent

Keynes famously compared investing to betting on the outcome of a beauty pageant.  With the RBA providing a little more transparency about their GDP forecasts in an appendix to the SOMP  there is now scope for a similar game — guessing how the RBA’s GDP forecast (for 2.8%y/y or approx. 0.6%q/q) is looking in light of incoming data.

Last week delivered a heavy blow for the RBA’s forecast. The construction work done data pretty much bolts right into the Private Dwelling Investment component of GDP, and it was a big miss. The RBA’s table tells us that they were expecting 5.9%y/y … the CWD release is consistent with growth of ~3%y/y. This line item is about 6% of GDP, so it shaved off ~15bps!

The CWD release also contained bad news about public activity. It fell around 6% (or 750mil), on the quarter. We can’t say for sure what the RBA expected, as their public demand forecast is the sum of public consumption and public capex — but we can put some bounds on things using their forecast of 4.2%y/y growth for public spending (consumption + capex).

Assuming that they went for a high public consumption number (say +1.25%q/q = 4.25%y/y), their forecast for public capex must have been around 3%q/q. Public building is only half of public capex, but it’s going to be very hard to hit +3%q/q for public capex when public building was down 6%q/q. In light of the construction data, I think it’s going to be more like -1%q/q. This line item is about 5% of GDP, so the 400bps miss means that it shaves off ~20bps from GDP.

The Private Capex release was a bit better than the RBA seems to have expected: I have private investment up about 0.5%q/q and 0.8%y/y, which is ~170bps better than the RBA’s -0.9%y/y forecast in the SOMP. That’s a gain of ~170bps in QoQ terms of an item that’s ~12% of GDP, so they picked up ~20bps from the Capex print.

In light of the above, my guess is that the RBA’s tracking estimate of Q4’18 GDP has fallen by around 20bps to 0.4%q/q (or 2.6%y/y). As best as i can tell, they are expecting a contribution of ~20bps from net exports, and roughly 10bps (or so) from inventories.

Assuming that NX and Inventories are ball-park, the focus should really be on the Government Finance Stats (released Tuesday). Given the size of the line item, the RBA’s chunky forecast (4.2%y/y, implies around 1.6%q/q), and the weak start from the CWD release, it’s this item that’s going to make or break the RBA’s Q4’18 GDP foreacst. And that’s important, as the RBA may be one downgrade away from an easing bias.