Even more real-time data

An economist colleague - oh all right, Kieran Murray of Sapere - sent me through some fascinating near-real-time economic data in Australia. It's from a partnership between illion, previously the Australian operations of Dun & Bradstreet, and Aussie economic consultants AlphaBeta.

Their very up to date data covers, among other things, consumer spending by region and by category, as well as indicators of personal or business financial distress. Here for example is the overall consumer spending graph. The distinction between the 'crisis' and 'stimulus' periods mainly reflects the impact of the A$750 cheques that went out from the end of March to over six million lower-income Australian households.

The sectoral breakdown of spend shows exactly how we've managed through lockdown: the biggest increases were on food delivery, online gambling, furniture and office (it's become evident from Zoom meetings that many folks, to start with, didn't have anything like a half-way equipped home office), home improvement (ditto), and alcohol and tobacco. While the bets and the booze don't reflect very well on how we've spent our enforced time at home, there's a small solace that the next largest increase was on pet care. Subscription TV was up too, though not as much as I'd have expected (11% more than in a 'normal' week).

This is excellent stuff. Yet again we've shown, if we put our minds to it, that we can actually get a much better handle on the real-time state of the economy than we've had before: if the Australian Treasury or others with skin in the game didn't know the likely scale and shape of the covid hit to retail sales, they do now.

Next thing is to keep this going on the other side of covid. As I've been banging on for a while (eg here and here), over the years we've had inadequate indicators of the cyclical state of the New Zealand economy: they've been too few and they've been too late. As we're now discovering in both New Zealand and Australia, there are treasure troves of near real-time data all over the place that can be used to fill in the gaps. Hopefully operators like illion and AlphaBeta will keep these data flows going but if not, and in any event, Stats NZ and their counterparts ought to be getting alongside these data providers and developing a suite of timely high-frequency indicators. Covid's been no time to be blundering in a statistical fog, but neither is anytime else.