Counterfactual Price Research Articles

Price level differences in the euro area: the case of Greece

Recent inflationary pressures have significantly affected household disposable incomes across Europe, with Greece being particularly impacted due to its low purchasing power. This study investigates the persistence and evolution of price level differences for fast-moving consumer goods in Greece compared to other euro area countries. It utilises the results of Dixon et al. (2023), who analysed price level differences across 41 product categories in ten euro area countries and found that the main factors contributing to price level differences include producer market competition, retail market structure, local costs and consumer habits. Building on these findings, we construct counterfactual prices and show that aligning Greece’s market structures and consumer behaviour patterns with the euro area average could significantly reduce prices (by 17 percentage points on average for the products with the highest share in total sales). The study also finds that although Greece has become cheaper in relative terms in recent years, it is still, on average, about 10% more expensive compared to euro area countries’ average. These results imply that there is scope for policy action, in particular, in areas that increase competition among producers, improve the structure of the retail market and enhance consumer economic literacy.

Managing the oil market under misinformation: A reasonable quest?

We examine the type and quality of information OPEC needs to successfully stabilize the oil market. Our analysis considers the impact of observational errors regarding market shocks as well as erroneous judgments of demand and supply elasticities. Actual prices resulting from OPEC's historical efforts to dampen volatility are compared to counterfactual prices that would have prevailed had OPEC remained passive. Despite the potentially confounding effect of misinformation, the elevated counterfactuals indicate that OPEC has managed to substantially decrease price volatility. Indeed, during the 2017–2021 OPEC+ period we estimate price volatility would have been up to 100% greater than actual without the actions of OPEC and its allies.

Full compensation and the volume effect: assessing different policy options

ABSTRACT According to the Damages Directive victims of infringements of competition law are entitled to full compensation. To achieve this goal overcharges, passing-on, and volume effects should play a role in the calculation of damages (for any purchaser who is not yet the final consumer). Whereas the Damages Directive promotes the passing-on defence for the defendants, it does not regulate volume effects in depth. The computation of these effects requires information on counterfactual prices and quantities. Since this information cannot be observed, the size of the volume effect tends to be uncertain. This paper discusses policy-options that aspire to bring compensation closer to full compensation, given uncertainty about the size of the volume effect. Based on the maximin-principle, a principle for decision-making under uncertainty, not allowing a passing-on defence appears to be an attractive option since this may lead to the minimization of the maximum gap between actual and full compensation.

Has retail competition reduced residential electricity prices in Texas?

We estimate the impact of introducing retail competition on retail electricity prices paid by residential consumers in Texas's two largest cities, Dallas and Houston. Using the synthetic control method to obtain the counterfactual prices, we find that retail competition raised average prices by $0.0112/kWh ($11.2/MWh) in the transition period from 2001 to 2006 and by $0.0134/kWh ($13.4/MWh) during the period of unfettered competition from 2007 to 2020. However, when the US wholesale natural gas prices are relatively low, actual retail electricity prices in areas opened to retail competition are close to the counterfactual prices that would have prevailed had retail competition not been introduced, as measured by the counterfactual prices estimated using the synthetic control method.

Behavioral changes in the housing market before and after the Covid-19 lockdown

We exploit unique Norwegian day-by-day transaction and hour-by-hour bidding logs data in order to examine how market participants reacted to the spreading news of Covid-19 in early March 2020, the lockdown on March 12, and the re-opening on April 20. We observe changes on the date of the lockdown in transaction volumes, sell-prediction spreads, exploitative bidding behavior, and seller confidence. However, when we compare observed price developments with our estimated counter-factual price developments, we find that about half of the total fall in prices had already occurred before the lockdown was implemented. The re-opening completely reverses the lockdown effect on prices. We show that voluntary behavioral changes, as well as lockdown and re-opening effects, are visible in various measures of social mobility, and that changes in daily news sentiment correlate with the abnormal price movements during this period.

The effect of the Canada–China canola trade dispute on canola prices

AbstractIn March 2019, China revoked the canola export licenses of two major Canadian exporters. We estimate the impact of these restrictions on Canadian canola prices. Using a vector error correction model to generate counterfactual prices, we estimate that between March 2019 and February 2020 canola prices were 3.6% lower than would have been expected in the absence of the import restrictions. We discuss the implications of our finding for both the ongoing negotiations between Canada and China and producer support in Canada.

A High-Dimensional Choice Model for Online Retailing

Online retailers are facing an increasing variety of product choices and diversified consumer decision journeys. To improve many operations decisions for online retailers, such as demand forecasting and price management, an important first step is to achieve a realistic understanding of the substitution patterns among a large number of products offered in the complex online environment. Although classic choice models offer an elegant framework for estimating substitution patterns among competing options, they have very limited applicability and performance in complex settings with many products. We provide a solution by developing a high-dimensional choice model that not only scales up more easily but also allows for flexible substitution patterns. We leverage consumer click-stream data, and combine econometric and machine learning (graphical lasso, in particular) methods to learn the substitution patterns among a large number of products. We show our model offers significantly better in- and out-of-sample demand forecasts in various synthetic datasets and in a real-world empirical setting. For example, in both settings, our model reduces the out-of-sample mean absolute percentage error (MAPE) by approximately 60% compared to classical models (i.e., the IID and the random coefficient Probit models). Our model can be used to enhance many business decisions such as assortment planning, inventory management and pricing decisions. In particular, we illustrate with a counter-factual pricing experiment that our model recommend better price points which increase total profit or revenue by 10% or above compared to classical models.

Residential property price differentials of waste plants: evidence from Beijing, China

ABSTRACTAs a primary source of urban pollution, waste plants release toxic gases and polluted waste water that can cause great harm to human health and contaminate water resources. The adverse impact of waste plants on environmental threaten the quality of life of surrounding residents which will be reflected in residential property values. In this study, we develop a hedonic price model to estimate the environmental externalities of waste plants based on a panel dataset for real-estate transactions in Beijing from 2011 to 2015. We apply fixed effects and a Heckman selection model to control for omitted variable bias and sample selection bias and then construct price counterfactuals for properties. The price differentials between expected and counterfactual prices are then calculated. Our empirical results suggest that waste plants negatively influence residential property prices. Additionally, the influences of other factors, such as school district, traffic convenience, and average housing area, on property prices are consistent with common sense, though they are weakened for properties within three kilometers of waste plants.

Milked and Feathered: The Regressive Welfare Effects of Canada’s Supply Management Regime: Reply

A comment by Doyon, Bergeron, and Tamini (2018) criticizes the approach and the results of a study by Cardwell, Lawley, and Xiang (2015) that quantifies the distributional effects that Canada’s supply management regime imposes on consumers. In this reply, we show that the main empirical result of Cardwell et al.—the degree of regressive distributional effects—is robust to alternative modelling choices and to alternative counterfactual price scenarios. We present new food price comparisons between Canada and the United States, showing that significant price premiums for supply-managed products persist under different exchange rates. Contrary to the results in Doyon et al., we find no evidence of systematic price premiums for non-supply-managed food products. Our new price comparisons highlight the shortcomings in the price comparisons in Doyon et al. and corroborate the results in Cardwell et al. Finally, we reject suggestions by Doyon et al. that our results are affected by research bias.

Nonparametric Welfare and Demand Analysis with Unobserved Individual Heterogeneity

In this paper, we combine elementary revealed preference principles and nonparametric estimation techniques in order to obtain nonparametric bounds on the distribution of the money metric utility over a population of heterogeneous households. The main benefit of our approach is that it is independent of any functional specification on the household utility functions, which means that our results are robust against parametric specification errors. We further demonstrate that our methodology can be used to establish bounds on the distribution of the demand function for counterfactual price regimes. In order to demonstrate the relevance of our approach, we illustrate our findings using a repeated cross-sectional household consumption data set.

What is the Appropriate Counterfactual When Estimating Effects of Multilateral Trade Policy Reform?

AbstractMultilateral trade reforms, such as may eventually emerge from the WTO's Doha Development Agenda (DDA), tend to be phased in over a decade or so after agreement is reached. Given the DDA's slow progress, that implementation may not be completed before the end of the next decade. Ex‐ante analysis of the DDA's possible effects thus requires first modelling the world economy to 2030 and, in that process, projecting what trade‐related policies might be by then without a DDA. Typically, modellers assume the counterfactual policy regime to be a ‘business‐as‐usual’ projection assuming the status quo. Yet we know developing country governments tend to switch from taxing to assisting farmers in the course of economic development. This paper shows the difference made by including political economy‐determined agricultural protection growth endogenously in the baseline projection. We reveal that difference by projecting the world economy to 2030 using the Global Trade Analysis Project (GTAP) model with those two alternative policy regimes and then simulating a move to global free trade (the maximum benefit from a multilateral trade reform) in each of those two cases. The welfare effects of removing the counterfactual price distortions in 2030 are shown to be much larger in the case where agricultural protection grows endogenously than in the case assuming no policy changes over the projection period. This suggests the traditional way of estimating effects of a multilateral agricultural trade agreement may considerably understate the potential welfare gains.

An empirical study of observational learning

This article provides an empirical examination of observational learning. Using data from an online market for music, I find that observational learning benefits consumers, producers of high‐quality music, and the online platform. I also study the role of pricing as a friction to the learning process by comparing outcomes under demand‐based pricing to counterfactual pricing schemes. I find that employing a fixed price (the industry standard) can hamper learning by reducing the incentive to experiment, resulting in less consumer surplus, but more expected revenue for the platform.

Democratic valuation (DV): Using majority voting principles to value public services

The now dominant survey approaches to value public services remain contentious. A common explanation is that stated preferences are generally difficult to measure. I argue that the key problems of the dominant survey approaches can be explained more specifically by their ambition to measure individuals' (maximum) willingness to pay (WTP). To estimate WTP researchers need to ask questions involving counterfactual prices, to specify doubtful implementation rules, and to aggregate preferences in a way that is inconsistent with political rights. The procedures engender problems with scenario credibility, incentive compatibility, adequate information provision, and democratic legitimacy. As an alternative to the standard stated preference paradigm I propose a democratic valuation approach (DV). DV measures the exchange value of public services in the political domain in terms of the aggregate valuation implied by the median preferred expenditure level for public services. The blueprint for survey design is the public finance referendum with multiple proposals. I suggest that empirically straightforward and democratically legitimate exchange values may be a more powerful input to public policy decisions than theoretically correct but empirically problematic estimates of mean willingness to pay.

Cannibalization or Market Expansion? The Impact of E-Books on Print Book Sales

E-book sales have experienced rapid growth in the past few years, yet their impact on print book sales remains unclear. This paper empirically investigates the cannibalization and market expansion effects of e-books. Using individual-level transaction data from 2008 to 2012, we estimate a dynamic structural model of consumer e-reader adoption and subsequent book purchases, including quantity, reading format (e-book or print book), and retailer choices in a number of book categories. We emphasize in the model the role of consumers’ self-selection into buying e-readers based on heterogeneous book tastes. We find that on average, 42% of e-book sales come from cannibalizing print book sales and 58% come from market expansion. Among the cannibalization effect, offline bookstores bear 53% of the cannibalization loss, while Amazon bears 32% and other online retailers bear 15%. We discuss how the impact of e-books would change under counterfactual pricing scenarios using the estimated demand system.

An Empirical Study of Observational Learning and Product Discovery

This paper is an empirical examination of observational learning. Using data from an online market for music, I find that observational learning benefits consumers, producers of high-quality music, and the online platform. I also study the role of pricing as a friction to the learning pro- cess by comparing outcomes under demand-based pricing to counterfactual pricing schemes. I find that employing a fixed price (the industry standard) can hamper learning by reducing the incentive to experiment, resulting in less consumer surplus, but more expected revenue for the platform.

Too Big to Be Efficient? The Impact of Implicit Subsidies on Estimates of Scale Economies for Banks

We examine whether “too‐big‐to‐fail” (TBTF) factors affect estimates of scale economies for large banks. From a standard model of bank production that does not control for any TBTF factors, we find evidence of scale economies for our sample of large banks. We then control for TBTF factors by using a measure of the “implicit subsidy” that emerges from a reduction in TBTF banks’ funding costs due to investor expectations of government support. We do this in two ways: first, we estimate scale economies from an augmented model of bank production that employs a proxy for the counterfactual price of debt that banks would face in the absence of any TBTF funding cost advantage; second, we estimate scale economies from a model of bank production that is estimated only for a sample of banks considered unlikely to be TBTF. After controlling for TBTF factors using either method, we no longer find evidence of scale economies for our sample of large banks. These results suggest that estimated scale economies for large banks are affected by TBTF factors.

Long-term Medicaid excess payments from alleged price manipulation of generic lorazepam.

Cost savings from the use of generic drugs versus brand-name drugs are well known. Both private and public prescription drug plans encourage the use of generic drugs through a variety of mechanisms. The magnitude of cost savings for a given generic drug is dependent on the degree to which the generic market is competitive. Should the competitive structure become compromised, higher prices and reduced cost savings may result. An alleged conspiracy between Mylan Laboratories and its active-ingredient suppliers in 1997 was associated with an increase in seller concentration in the generic lorazepam market. The Federal Trade Commission (FTC) alleged that Mylan raised costs to consumers by $120 million because of price increases for generic lorazepam from March through December 1998 and for generic clorazepate from January through December 1998. In November 2002, a settlement with Mylan was approved by the FTC, and a federal district court required Mylan to pay $147 million, including $28.2 million to state agencies including Medicaid. To (a) describe the seller concentration in the national Medicaid generic lorazepam market over a 19-year period from January 1991 through December 2009, (b) estimate the excess payments for generic lorazepam by Medicaid between 1998 and 2009, and (c) investigate potentially increased utilization and prices of 2 substitute pharmaceuticals: branded lorazepam (Ativan) and generic alprazolam (another widely used intermediate-acting benzodiazepine). Using Medicaid State Drug Utilization Data from the Centers for Medicare Medicaid Services, we calculated the 4-firm concentration ratio (CR₄) and the Herfindahl-Hirschman Index (HHI) for the Medicaid generic lorazepam market, along with pre-rebate reimbursement for pharmacy claims, number of claims (utilization), and average pre-rebate reimbursement per claim (average "price") for generic lorazepam, from 1991 through 2009. Medicaid's excess payments were estimated under 2 different assumptions regarding what the average generic lorazepam price would have been in the absence of the alleged conspiracy. To find counterfactual prices, the average per-claim reimbursement for lorazepam for the 4 quarters prior to the alleged conspiracy, $6.80, was inflated using (a) the quarterly change in the average per-claim reimbursement for generic alprazolam and (b) the Consumer Price Index (CPI) for all urban consumers, all goods. Potential impact of the alleged conspiracy on the branded lorazepam and generic alprazolam markets was investigated. The average pre-rebate reimbursements per claim for generic lorazepam were $10.25, $23.12, and $8.48 in 1991, 1998, and 2009, respectively. For the same 3 years, CR₄ = 52.80, 76.02, and 86.74, while HHI = 905.71, 2,166.25, and 2,233.36. Medicaid's excess payments from 1998-2009 were estimated at approximately $625-$657 million. The data also suggest the possibility of small impacts on the utilization of branded lorazepam and the price of generic alprazolam. Prior to the alleged conspiracy in 1997, average pre-rebate reimbursement per claim for generic lorazepam was declining, while seller concentration was rising. After a jump in average payment per claim in the years immediately following the alleged conspiracy, prices have gradually returned to their pre-1998 levels. However, the generic lorazepam market was more concentrated in 2009 than prior to the alleged conspiracy.

Crisis, sustainability of electricity prices and state interventions in Argentina

This article examines the sustainability of wholesale power prices following the 2001 economic crisis in Argentina. We depart from the contractual disputes between the multinational power generating companies and the state of Argentina, arising from price and investment interventions after the economic crisis. Then, we move on to argue that these sector-specific measures were inevitable once some of the key macroeconomic policies such as the currency board system were discarded. Using an error correction method and estimating a counterfactual price series, we demonstrate that in the absence of changes in the price determination method and direct public investment, the level and the volatility of power prices would have been much higher with further knock on effects on the rest of the economy. Copyright 2012 The Author 2011. Published by Oxford University Press on behalf of Associazione ICC. All rights reserved., Oxford University Press.

Commentary—Do Bids Equal Values on eBay?

We argue that the Zeithammer and Adams paper [Zeithammer, R., C. Adams. 2010. The sealed-bid abstraction in online auctions. Marketing Sci. 29(6) 964–987] successfully documents consistent patterns in eBay bidding data that cast doubt on the common assumption that bidders in such auctions follow a “bid = value” strategy. These anomalies lend support to the authors' alternative model in which some bidders bid reactively and consequently bid below their valuation most of the time. The consistency of the authors' findings as well as the ability of their alternative explanation to account for all of their test results lends great support to their thesis. However, we think that several of their empirical tests examine ancillary assumptions about bidder behavior and do not test the bid = value assumption directly. Furthermore, although their reduced-form model incorporating “reactive” bidders is a good first attempt at expanding the canonical framework, we worry that their counterfactual pricing analysis using the reactive model is suspect because the parameters they estimate are not structural. Overall, the Zeithammer and Adams paper is a carefully argued critique of empirical methods used to study online auctions and provides valuable ideas to improve on these methods.

Exchange Rates and Cash Flows in Differentiated Product Industries: A Simulation Approach

ABSTRACTHow do exchange rate changes impact firms' cash flows? We extend a simulation method developed in industrial organization to answer this question. We use prices, quantities, and product characteristics for differentiated products, coupled with a discrete choice framework and an assumption of price competition, to estimate marginal costs for all producers. Using a Monte Carlo approach we generate counterfactual prices and profits for different levels of exchange rates. We illustrate the method using the market for bottled water. Our results stress that even in a relatively simple market such as this one, different brands face very different exchange rate risks.