# Anshuman Sahoo

• Research Fellow, Steyer-Taylor Center for Energy Policy and Finance
• Knight Managemnet Center
• 655 Knight Way
• Stanford, CA 94305
• asahoo 'at' stanford 'dot' edu

Research interests: Energy, sustainability, and public policy

## Overview of Research

Published Papers
The two papers in Energy Policy below study the effectiveness of a Domestic Content Requirement (DCR). The first provides an ex-ante analysis, while the second provides an ex-post evaluation of its performance in the first phase of the Jawaharlal Nehru National Mission.

The Effectiveness of Domestic Content Criteria in India's Solar Mission
With Gireesh Shrimali
Abstract: Often, a goal of renewable energy policies is the development of domestic renewable energy technology manufacturing capacity. The Jawaharlal Nehru National Solar Mission (NSM) in India is an example; besides targeting an installation of 20GW of grid-tied solar power capacity, it includes a domestic content requirement (DCR) to strengthen a solar photovoltaic manufacturing base. We ask whether the DCR of the NSM will be effective in ensuring the global competitiveness of the beneficiary sector. Our analysis reveals three observations that indicate this outcome is unlikely: (1) the manufacturing base has become less competitive over time, (2) developers may be favoring thin-film technology, thereby bypassing the DCR, which applies specifically to crystalline silicon cells and modules, and (3) gaps in the Indian innovation system are likely to prevent a return to competitiveness by solar photovoltaic manufacturers. In particular, a comparison with the Chinese innovation system indicates shortcomings in the Indian innovation system of R&D capabilities, coordination of resource provision and complementary industrial strengths. Given these observations, we suggest that policymakers remove the solar photovoltaic DCR from the NSM.
Article in Energy Policy via ScienceDirect
Working paper version

Has India's Solar Mission increased the deployment of domestically produced solar modules?
With Gireesh Shrimali
Abstract: The Jawaharlal Nehru National Solar Mission (JNNSM), India's flagship policy for solar energy deployment, includes an increasingly strict Domestic Content Requirement (DCR) intended to promote the domestic crystalline photovoltaic solar industry. We examine the impact of the JNNSM DCR on the utilization of domestic modules. Using a plant level database of approximately 250 plants, we show that the first and weaker version of the policy in the JNNSM Batch I promoted the use of domestic crystalline silicon technology in place of foreign crystalline silicon technology. However, the second and stricter version of the policy in the JNNSM Batch II has not been as effective: it has promoted the use of foreign thin film modules in place of foreign crystalline silicon modules. This analysis shows that upon tightening the DCR requirements between JNNSM Batches I and II, Indian policymakers allowed for leakage to foreign thin film modules. DCR policies need to be comprehensive to ensure that the intended goal of using only domestic content is realized; in doing so, however, policymakers should carefully assess the welfare impacts of such comprehensive restrictions.
Article in Energy Policy via ScienceDirect

Paper under Review

Time of Day Pricing and the Levelized Cost of Intermittent Electricity Generation
With Stefan Reichelstein
Abstract: An important characteristic of most renewable energy sources is their intemittent pattern of electricity generation. Yet, intermittency is usually ignored in life-cycle cost calculations intended to assess the competitiveness of electric power from renewable as opposed to dispatchable energy sources, such as fossil fuels. This paper demonstrates that for intermittent renewable power sources a traditional life-cycle cost calculation should be appended by a correction factor which we term the Co-Variation coefficient. It captures any synergies, or complementarities, between the time-varying patterns of power generation and pricing. We estimate the Co-Variation coefficient for specific settings in the western United States. Our estimates imply that the benchmark of cost competitiveness for solar PV power is 10-15% lower than average life-cycle costs have suggested. In contrast, the generation pattern of wind power exhibits complementarities with electricity pricing schedules, yielding a cost competitiveness assessment 10-15% above that suggested by traditional calculations.
Working paper version
Revise and resubmit, Energy Economics

KiOR - The Quest for Cellulosic Biofuels
With Stefan Rosenthal and Sara Reichelstein
Case E427, Stanford GSB Case Series, 2013

Teaching Note for KiOR - The Quest for Cellulosic Biofuels
With Stefan Reichelstein
Teaching Note E427-TN, Stanford GSB Case Series, 2013
Case and note available from Harvard Business School Press

Work in progress

Cost- and Price Dynamics of Solar PV Modules
With Stefan Reichelstein
Details: Solar photovoltaic (PV) module prices have adhered remarkably well to an 80\% learning curve, but recent price declines have been even steeper than predicted. Analysts have questioned whether these price declines reflect underlying production cost reductions or excess manufacturing capacity. To examine this issue, we develop methods to estimate production costs from public financial accounting data and derive prices that would emerge in a long-run equilibrium. These prices follow from a dynamic model of competition in which each firm makes periodic additions to capacity and subsequently chooses output levels. The resulting market prices in equilibrium, which we term \textit{Economically Sustainable Prices (ESP)}, correspond to the long-run marginal cost of producing modules. Applying this framework, we demonstrate that recent dramatic module price reductions were significantly attributable to excess capacity. We also estimate how production costs for modules have changed as a function of time and experience. This estimate allows us to extrapolate a trajectory of future equilibrium prices to which actual market prices should converge over time. The forecast price trend reflects a rate of cost declines greater than that embedded in the 80\% learning curve.
Presentation of an earlier version to the Stanford Energy Seminar

Managerial Flexibility in Measures of Levelized Cost
With John Bistline, Stephen Comello and Stefan Reichelstein
Details: Many irreversible long-run capital investment options entail opportunities for managers to respond flexibly to changes in the economic environment. However, common levelized cost measures designed to guide decision-making implicitly assume that the values of random economic variables are known with certainty when investment decisions are made. This assumption implies, often incorrectly, that managerial flexibility carries zero value. We improve levelized cost measures by deriving an expansion that accounts for both uncertainties in relevant variables and the value of managerial flexibility in responding to them. We apply our method to quantify the value of flexibility in an example decision problem in which an operator of a natural gas power plant evaluates whether to invest in carbon capture and sequestration (CCS) capabilities.

Negative Dividends: Internality Losses can Outweigh Externality Gains
With Nik Sawe
Details: The Energy Star program, which highlights energy efficient goods with a label, has been credited with increasing the uptake of such products and thereby reducing the external costs of energy consumption. Nonetheless, the social value of the program is ambiguous if, in its absence, consumers both over-value and under-value the energy consumption attribute of alternative goods, relative to their economic preferences. To guarantee an increase in consumer welfare in the presence of such heterogeneity, the Energy Star must prompt some individuals to decrease their valuation of the energy consumption attribute and others, to increase it. If unable to appropriately exert both effects, the Energy Star could yield negative dividends" by inducing changes in consumer behavior that imply aggregate losses in individual-level welfare that exceed gains from externality reductions. We develop a method to quantify the impact of the Energy Star on individual-level decision-making behavior and welfare. We illustrate the valuation methodology with unique data from a stated choice experiment involving light bulbs. Our example demonstrates the potential for negative dividends and that the social value of programs such as the Energy Star depends on the choice set available to consumers.