Thread by @TobyMPhillips

I did a quick spot of analysis on the UK's Eat Out to Help Out scheme... what was the impact of 50% discounts on dine-in meals?
Posted today at @ConversationEDU @ConversationUK. A few key points... 👇🏼

theconversation.com/eat-out-to-hel…

#CovidUK

1/ the policy gave people a 50% subsidy (up to £10) on meals on Mon-Wed during August. This *definitely* enticed more people to dine out – almost 2x previous Aug 2019 levels on relevant days.

(this is data from OpenTable)

2/ after the policy finished, things went straight back to where we would have expected based on the pre-August trend.
No change in behaviour seems to have persisted into September.

3/ looking at Google data, there is a slight (but smaller) effect in terms of overall outings to retail, recreation and hospitality establishments. So it didn't lead to a massively more outings to the High Street.

Again, in September, retail mobility dropped back

4/ So the scheme was a qualified success: it definitely got cash into the hands of struggling hospitality businesses.

But it didn't seem to spill over much. Not to other days of the week. Not to September. And not to other businesses (at least not in footfall).

5/ but the scheme has consequences! The rate of covid cases has been rising across the UK over the end of August and into September.
We can't say for sure if this is *caused* by Eat Out to Help Out, but it certainly didn't encourage people to stay home.

6/ all-up, the lessons are clear: it is fine to support businesses, but this should be designed to work for the long-haul, and designed to work *during an ongoing pandemic*.
Spread people out; don't concentrate them.

CC: a few people who have been writing about the recent cases uptick and Eat Out to Help Out scheme @alicemhancock @PhilAldrick @Bordeaux8 @AndyBounds @_ManishPandey @patrickjbutler @ChrisGiles_ @matthabusby @SYTKevinRogers @devisridhar