Representing Electricity: Economics, Physics, and Computing

• In Event: Precision and Uncertainty in a World of Data - II

Abstract

What kind of expertise does it take to trade electricity in contemporary markets? Depending on where one stands, the answer may vary from Kirchhoff’s electric current laws, to fluency in programming languages, to marginal cost theory. This paper explores the heterogenous nature of trading expertise in a quirky commodity—one that needs to be in physical balance at all times in order for its infrastructure (i.e., the electric grid) to stay intact. In the United States, as elsewhere, the electricity industry was considered as the quintessential “natural monopoly” for most of its existence. To reverse that reputation, in the 1980s, groups of electrical engineers and economists worked to blend electricity physics with marginal cost theory; they invented pricing algorithms that followed neoclassical economics and the laws of electricity transmission concurrently. Once codified into algorithms, electricity-specific knowledge was supposed to cease to be essential to the daily work of traders, but algorithms’ connection to electricity’s transmission has to be monitored, even remade, continuously. In this paper, I report on my fieldwork at an electricity trading company in Boston, Massachusetts. To sift profit opportunities through an avalanche of data before daily deadlines, traders need codes that render electricity data computationally standard. While such codifications allow programmers to enter the field of electricity trade with little electricity-specific expertise, the balance of expertise at play remains fragile. Electricity complicates Michel Callon’s theory of the performativity of economics—it illustrates that “framing” a commodity is seldom possible by economics expertise alone.

Author

• Canay Ozden-Schilling, Johns Hopkins University