Working Paper: NBER ID: w19510
Authors: Ben Gilbert; Joshua S. Graff Zivin
Abstract: Digital tracking and the proliferation of automated payments have made intermittent billing more commonplace, and the frequency at which consumers receive price, quantity, or total expenditure signals may distort their choices. This category of goods has expanded from household utilities, toll road access and software downloads to standard consumption goods paid by credit card or other "bill-me-later"-type systems. Yet we know surprisingly little about how these payment patterns affect decisions. This paper exploits hourly household electricity consumption data collected by "smart" electricity meters to examine dynamic consumer behavior under intermittent expenditure signals. Households reduce consumption by 0.6% to 1% following receipt of an electricity bill, but the response varies considerably by household type and season. Our results also suggest that spending "reminders" can reduce peak demand, particularly during summer months. We discuss the implications for energy policy when intermittent billing combined with inattention induces consumption cycles.
Keywords: dynamic salience; intermittent billing; electricity consumption; smart meters
JEL Codes: D03; Q4
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Reduction in Electricity Consumption (Q41) | Reduction of Air Conditioning Use During Peak Hours (L97) |
| Electricity Bill Receipt (L94) | Reduction of Air Conditioning Use During Peak Hours (L97) |
| Reduction of Air Conditioning Use During Peak Hours (L97) | Adjusted Consumption Patterns Temporarily (D12) |
| Electricity Bill Receipt (L94) | Reduction in Electricity Consumption (Q41) |