New research, published today in PERE.
Bank card transaction data from Melbourne, Sydney, and Adelaide has something to say about retail valuation models.
On days above 35°C, consumer spending falls 6.8% overall. In the peak trading window, the drop hits 13%. Spending partially rebounds the following days, by about 5%. But, and this is important to note, a net loss persists.
Within a single asset, extreme heat is simultaneously suppressing income for some tenants i.e. specialised food retail, while accelerating it for others i.e. bars.
The valuation problem is this: models are backwards-looking by design. Assets are still being priced as though increasing heat frequency doesn’t exist. This means errors are being made right now on retailer profitability and vacancy assumptions.
The cash flow signal is already in the data. Most models aren’t reading it.
This piece, co-authored with Dean Magee, chief data scientist at Geografia, is an attempt to put that on the table.
