Skip to main content
Honesty box shortfall turns on the change in the tin

Honesty box shortfall turns on the change in the tin

A new Slop University research poster randomises the float and the unit pricing at 58 roadside honesty-box stalls, and finds a district listing scheme's honesty index tracking whether a passing driver could make exact change.

A randomised crossover trial at 58 farm-gate stalls moves payment shortfall 17.5 points on float and unit pricing alone

A driver who means to pay and finds they cannot is not a dishonest driver. For eleven weeks that was the only difference we introduced, and it moved the figure the districts publish.

— Associate Professor Kwame Lindqvist, Director, Master of Priority Studies

No Change in the Tin, a research poster released today through the Office of Research Outputs, randomised two things at 58 farm-gate stalls across four shire districts: whether a float of coins sat in the tin, and whether the day’s unit price came to a whole dollar. Payment shortfall ran at 22.6 per cent of the value taken in the weeks with neither and 5.1 per cent in the weeks with both. Once float and pricing are accounted for, which stall it was explains 7 per cent of what is left.

The same takings feed the honesty index by which the four districts list stalls on a tourist drive. Twenty-one of the twenty-four stalls currently listed as high trust keep a permanent float; the three that do not price every line in whole dollars.

“Change is a fixed cost, and a stall that cannot leave eighteen dollars out overnight is the stall the index drops,” said Dr Fenna Okoro, Senior Lecturer and Convenor of the Horizon Register. “A listing decision is being made on a condition nobody set out to test.”

The work extends the School of Emergent Priorities’ longer interest in what a listing starts to select for once it exists, and in the conditions a ranking quietly requires of the things it ranks. The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.pfhong.