Search

This paper presents a structural discrete choice model with social influence for large-scale social networks. The model is based on an incomplete information game and permits individual-specific parameters of consumers. It is challenging to apply this type of models to real-life scenarios for two reasons: (1) The computation of the Bayesian–Nash equilibrium is highly demanding; and (2) the identification of social influence requires the use of excluded variables that are oftentimes unavailable. To address these challenges, we derive the unique equilibrium conditions of the game, which allow us to employ a stochastic Bayesian estimation procedure that is scalable to large social networks. To facilitate the identification, we utilize community-detection algorithms to divide the network into different groups that, in turn, can be used to construct excluded variables. We validate the proposed structural model with the login decisions of more than 25,000 users of an online social game. Importantly, this data set also contains promotions that were exogenously determined and targeted to only a subgroup of consumers. This information allows us to perform exogeneity tests to validate our identification strategy using community-detection algorithms. Finally, we demonstrate the managerial usefulness of the proposed methodology for improving the strategies of targeting influential consumers in large social networks. This paper was accepted by Matthew Shum, marketing.

Shop-floor employees play a key role in manufacturing innovation. In some companies, up to 75% of all productivity gains are the result of bottom-up employee ideas. In this paper, we examine how employee interplant assignments—short problem-solving jobs at other manufacturing plants within the same firm—influence employee-driven manufacturing innovation. Using unique idea-level data from a large European car parts manufacturer, we show that interplant assignments significantly increase the value of employees’ improvement ideas due to the short-term transfer of production knowledge and long-term employee learning. Both effects are amplified by assignments to plants that have high functional overlap (i.e., plants producing similar products using similar processes and machinery). One implication is that, for the purpose of employee-driven manufacturing innovation, assignments between peripheral plants with high functional overlap can be more effective than assignments to and from central plants. These findings are robust to several econometric tests. Our study provides novel and detailed empirical evidence of manufacturing innovation, and goes beyond previous research on the learning curve (learning by doing) by investigating how interplant assignments affect the value of employees’ improvement ideas (learning by moving). This paper was accepted by Charles J. Corbett, operations management.

This paper analyzes price comparison platforms’ equilibrium design of their search environments, namely the order under which firms are displayed (having a prominent firm or listing firms randomly) and how much price complexity firms are able to use. It is found that (1) the possibility to obfuscate amplifies firms’ willingness to pay for the prominent position; (2) when platforms sell prominence, they optimally allow for more obfuscation than they would if they were to display firms randomly; and (3) if platforms are sufficiently differentiated, they find it optimal to sell prominence. Thus, platforms exploit their market power over consumers by implementing a prominent position and allowing for large levels of obfuscation. This paper also finds that in equilibrium, there is tension between platforms and firms regarding how much price complexity is used: firms would like to use even more obfuscation than what the platform allows, so the platform must monitor firms’ prices to make sure that they are not excessively complex. This paper was accepted by Juanjuan Zhang, marketing.

Firms’ innovative activities can be sensitive to public policies that affect the availability of capital. In this paper, we investigate the effects of regional and temporal variation in U.S. personal bankruptcy laws on firms’ innovative activities. We find that bankruptcy laws that provide stronger debtor protection decrease the number of patents produced by small firms. Stronger debtor protection also decreases the average quality, and variance in quality, of firms’ patents. We find evidence that the negative effect of stronger debtor protection on experimentation and innovation may be due to the decreased availability of external financing in response to stronger debtor rights, an effect amplified in industries with a high dependence on external financing. Hence, while it is typically assumed that stronger debtor protection encourages innovation by reducing the cost of failure for innovators, we show that it can instead dampen innovative activities by tightening the availability of external financing to innovative firms. This paper was accepted by David Hsu, entrepreneurship and innovation.

The recent influx of migrants and refugees into Europe and elsewhere raises questions as to whether migrant behavior reflects cultural predispositions and whether assimilation through exposure to host institutions can be expected. The paper focuses on financial behavior and uses high-quality administrative data on migrants and refugees to Sweden. It uncovers differences across cultural groups in how behavior relates to household characteristics, and shows that differences diminish with exposure to host country institutions, even for large cultural distances. Interestingly, robust cultural classification of European countries based on genetic distance or on Hofstede’s cultural dimensions fails to identify a single “southern” culture but points to a “northern” culture. Our results also have implications for the potential of European institutional harmonization, exogenously imposed during the fiscal crisis, to alleviate cultural differences in financial behavior. This paper was accepted by Lauren Cohen, finance.

We examine how freemium platforms can design social referral programs to encourage growth and engagement without sacrificing revenue. On the one hand, social referral programs generate new referrals from users who would not have paid for the premium features. On the other hand, they also attract new referrals from users who would have paid but prefer to invite others, resulting in more referrals but fewer paying users. We use data from a large-scale randomized field experiment in an online dating platform to assess the effects of adding referrals programs to freemium platforms and changing the referral requirements on users’ behavior, namely, on their decisions to invite, pay, and engage with the platform. We find that introducing referral programs in freemium platforms can significantly contribute to increasing the number of referrals at the expense of revenue. Platforms can avoid the loss in revenue by reserving some premium features exclusively for paying users. We also find that increasing referral requirements in social referral programs can work as a double-edged sword. Increasing the referral threshold results in more referrals and higher total revenue. Yet these benefits appear to come at a cost. Users become less engaged, decreasing the value of the platform for all users. We explore two mechanisms that help to explain the differences in users’ social engagement. Finally, and contrary to prior findings, we find that the quality of the referrals is not affected by the referral requirements. We discuss the theoretical and practical implications of our research. This paper was accepted by Chris Forman, information systems. Funding: This work was funded by Fundac¸ão para a Ciência e a Tecnologia [UID/ ECO/00124/2019, UIDB/00124/2020, UIDP/00124/2020, and Social Sciences DataLab - PINFRA/22209/2016], POR Lisboa, and POR Norte [Social Sciences DataLab, PINFRA/22209/2016]. Supplemental Material: The online appendix and data are available at https://doi.org/10.1287/mnsc.2022.4301 .

Providing broadband to schools can be an effective way to foster household Internet adoption in neighboring areas. On the one hand, the infrastructure put into place to meet schools’ needs can also serve households. On the other hand, students get acquainted with the Internet at school and signal its usefulness to adults at home who, consequently, can be more likely to adopt it. In this paper, we model the roles that broadband use at school and Internet adoption in neighboring households play in the decision to adopt the Internet at home and measure their effects empirically. We use data from Portugal between 2006 and 2009 on household Internet penetration and on how much schools use broadband. We use two different sets of instruments for the schools’ broadband use to alleviate endogeneity concerns. Both approaches yield similar results. We find that broadband use at school leads to higher levels of Internet penetration in neighboring households. Broadband use in schools was responsible for a year-over-year increase of 3.5 percentage points on Internet penetration in households with children. Across our data set this effect accounts for about 17% of the increase in home Internet adoption. We also find evidence of regional spillovers in Internet adoption across households. These were roughly responsible for an increase of 2.1 percentage points in Internet penetration or 38% of the total increase in household Internet penetration between 2006 and 2009. These results show that wiring schools with broadband is an effective policy to lower the barriers for Internet adoption at home and as such contributes to accelerating the pace of broadband diffusion. This paper was accepted by Lorin Hitt, information systems.

This paper reports the results from a controlled field experiment designed to investigate the causal effect of unannounced, public recognition on employee performance. We hired more than 300 employees to work on a three-hour data-entry task. In a random sample of work groups, workers unexpectedly received recognition after two hours of work. We find that recognition increases subsequent performance substantially, and particularly when recognition is exclusively provided to the best performers. Remarkably, workers who did not receive recognition are mainly responsible for this performance increase. Our results are consistent with workers having a preference for conformity and being reciprocal at the same time. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mnsc.2015.2291 . This paper was accepted by John List, behavioral economics.

This paper explores how loss-framed incentives affect behavior in a multitasking environment in which participants have more than one way of recovering (expected) losses. In a real-effort laboratory experiment, we offer participants task incentives that are framed as either a reward (gain) or penalty (loss). We study their responses along three dimensions: performance in the incentivized task, theft, and voluntary provision of help. We find that framing incentives as a penalty rather than as a reward does not significantly improve task performance, but it increases theft and leads to a small and insignificant reduction in the share of participants willing to help the experimenter. Secondary analyses based on our theoretical framework help us pin down the mechanism at play and suggest that loss aversion drives participants’ response. Our findings have important implications for incentive design in practice. David Jimenez-Gomez is grateful for the funding from the Ministry of Science and Innovation and the European Regional Development Fund [Grant PID2019-107081GB-I00], from the Ministry of Science and Innovation through the project “Nudging applied to the improvement of regulation” [Grant RED2018-102761-T], and from the Valencian Community through the Prometeo program [Grant PROMETEO/2021/073].

The general data protection regulation (GDPR) represents a dramatic shift in global privacy regulation. We focus on GDPR’s enhanced consumer consent requirements that aim to provide transparent and active elicitation of data allowances. We evaluate the effect of enhanced consent on consumer opt-in behavior and on firm behavior and outcomes after consent is solicited. Utilizing an experiment at a large telecommunications provider with operations in Europe, we find that opt-in for different data types and uses increased once GDPR-compliant consent was elicited. However, consumers did not uniformly increase data allowances and continued to generally restrict permissions for more sensitive or tangential uses of their personal information. We also find that sales, the efficacy of marketing communications, and contractual lock-in increased after consumers provided new data allowances. Additional analysis suggests that these gains to the firm emerged because new data allowances enabled them to increase their use of targeted marketing for households that were amenable to these marketing efforts. These results have significant implications for firms and policymakers and suggest that enhanced consent provided via GDPR may be effective for increasing consumer privacy protection while also allowing firms reliant on consumers’ personal information to improve outcomes. This paper was accepted by Chris Forman, information systems.

We propose an optimal currency hedging strategy for global equity investors using currency value, carry, and momentum to proxy for expected currency returns. A benchmark risk constraint ensures the overlay closely mimics a fully hedged portfolio. We compare this with naïve and alternative hedges in a demanding out-of-sample test, with transaction and rebalancing costs and margin requirements. Other hedging methods generally reduce risk but at a cost. Some tend to short currencies with high returns and all incur substantial costs with frictions, mostly margin requirements and equity rebalancing costs. The proposed strategy uses predictable returns to reduce this cost. It produces a statistically significant 17% gain in Sharpe ratio and an annualized Jensen-α of 0.93% versus a fully hedged benchmark. Notably, most of the implementation costs of the strategy would be incurred by the benchmark anyway. This reduces its marginal cost and highlights a specific synergy of integrating hedging with speculation. This paper was accepted by Gustavo Manso, finance.

This paper is on production planning and inventory control in systems where manufacturing and remanufacturing operations occur simultaneously. Typical for these hybridsystems is, that both the output of the manufacturing process and the output of the remanufacturing process can be used to fulfill customer demands. Here, we consider a relatively simple hybrid system, related to a single component durable product. For this system, we present a methodology to analyse a PUSH control strategy (in which all returned products are remanufactured as early as possible) and a PULL control strategy (in which all returned products are remanufactured as late as is convenient). The main contributions of this paper are (i) to compare traditional systems without remanufacturing to PUSH and to PULL controlled systems with remanufacturing, and (ii) to derive managerial insights into the inventory related effects of remanufacturing.

We study the empirical relevance of the participation externality between liquidity suppliers (makers) and demanders (takers), i.e., whether liquidity demand attracts or reduces liquidity supply, and vice versa. We use exogenous shocks to exchange fees and technology as experiments to identify cross-sided complementarities between liquidity suppliers and demanders in the U.S. equity market. We find that the externality is large and positive, on average. However, the externality is negative in periods of high adverse selection. We quantify the economic significance of the externality by evaluating an exchange’s revenue after a fee change. The Internet appendix is available at https://doi.org/10.1287/mnsc.2016.2658 . This paper was accepted by Lauren Cohen, finance.

We explore how individual equilibrium effort in tournaments varies with the number of contestants. The probability of winning a tournament depends on both effort and luck, and we show that the distribution of the luck component is critical in determining individual equilibrium effort. Our theory predicts that equilibrium effort is an increasing (decreasing) function of the number of contestants if there is considerable (little) mass on favorable draws. We test our theory using both laboratory and field experiments, and find substantial support for our theory in both settings. This paper was accepted by Teck Ho, behavioral economics.

Many online markets encourage traders to make good after an unsatisfactory transaction by offering the opportunity to withdraw negative reputational feedback in a dispute resolution phase. Motivated by field evidence and guided by theoretical considerations, we use laboratory markets with two-sided moral hazard to show that this option, contrary to the intended purpose, produces an escalation of dispute. The mutual feedback withdrawal option creates an incentive to leave negative feedback, independent of the opponent’s behavior, to improve one’s bargaining position in the dispute resolution phase. This leads to distorted reputation information and less trust and trustworthiness in the trading phase. Buyers who refuse to give feedback strategically, even when it comes at a personal cost, mitigate the detrimental impact. It is also mitigated in markets with one-sided moral hazard and a unilateral feedback withdrawal option. The supplementary appendix is available at https://doi.org/10.1287/mnsc.2017.2802 . This paper was accepted by Uri Gneezy, behavioral economics.

We partner with a major multinational telecommunications provider to analyze the effect of subscription video-on-demand (SVoD) services on digital piracy. For a period of 45 consecutive days, a group of randomly selected households who used BitTorrent in the past were gifted with a bundle of TV channels with movies and TV shows that could be streamed as in SVoD. We find that, on average, households that received the gift increased overall TV consumption by 4.6% and reduced Internet downloads and uploads by 4.2% and 4.5%, respectively. However, and also on average, treated households did not change their likelihood of using BitTorrent during the experiment. Our findings are heterogeneous across households and are mediated by the fit between the preferences of households in our sample for movies and the content available as part of the gifted channels. Households with preferences aligned with the gifted content reduced their probability of using BitTorrent during the experiment by 18% and decreased their amount of upload traffic by 45%. We also show using simulation that the size of the SVoD catalog and licensing window restrictions limit significantly the ability of content providers to match SVoD offerings to the preferences of BitTorrent users. Finally, we estimate that households in our sample are willing to pay at most $3.25 USD per month to access a SVoD catalog as large as Netflix’s in the United States. Together, our results show that, as a stand-alone strategy, using legal SVoD to curtail piracy will require, at the minimum, offering content much earlier and at much lower prices than those currently offered in the marketplace, changes that are likely to reduce industry revenue and that may damage overall incentives to produce new content while, at the same time, curbing only a small share of piracy. The online appendix is available at https://doi.org/10.1287/mnsc.2017.2875 . This paper was accepted by Chris Forman, information systems.

Using a novel experimental paradigm, we explore how the experience of generating an idea and the possibility that another investor might adopt a rejected investment opportunity, bias the investment decisions of innovator and imitator entrepreneurs. We find that individuals who generate a business idea form biased evaluations of the economic potential of ideas, be it their own idea or somebody else’s idea. On the one hand, they are overconfident about the value of, and overly likely to invest in, their own idea. On the other hand, when investing in another person’s idea, even if it is not competing with their own idea, they are underconfident about the value of, and insufficiently likely to invest in, the idea. Surprisingly, we find that entrepreneurial experience exacerbates this pattern of over- and underconfidence. In addition, we find that the threat that another investor can appropriate a declined investment opportunity increases willingness to invest. We propose a theoretical account to explain the observed pattern of over- and underconfidence in imitative and innovative entrepreneurship. Our findings challenge the traditional account that lowering the cost of imitation has a disincentive effect on the investment decisions of pioneer entrepreneurs and provide evidence that a more lenient appropriability regime may, unexpectedly, have positive effects on entrepreneurship. Our findings also identify new psychological mechanisms that can play a role in important phenomena such as the emergence of spin-offs and rush to market entry. Data, as supplemental material, are available at https://doi.org/10.1287/mnsc.2016.2566 . This paper was accepted by Uri Gneezy, behavioral economics.

We investigate Morningstar’s new qualitative, forward-looking analyst ratings, which reflect independent analysts’ expectations of a fund’s future performance. We find relatively higher flows to funds receiving higher ratings, suggesting that the average investor values the analyst’s subjective views when allocating their wealth. Performance tests show that investors would have earned significantly higher returns over our sample period by investing in funds with the highest analyst conviction. These results suggest that independent research that expands the information set to include qualitative elements may help investors make better investment allocation decisions. This paper was accepted by Lauren Cohen, finance.

Using a theorem showing that matching probabilities of ambiguous events can capture ambiguity attitudes, we introduce a tractable method for measuring ambiguity attitudes and apply it in a large representative sample. In addition to ambiguity aversion, we confirm an ambiguity component recently found in laboratory studies: a-insensitivity, the tendency to treat subjective likelihoods as 50-50, thus overweighting extreme events. Our ambiguity measurements are associated with real economic decisions; specifically, a-insensitivity is negatively related to stock market participation. Ambiguity aversion is also negatively related to stock market participation, but only for subjects who perceive stock returns as highly ambiguous. This paper was accepted by James Smith, decision analysis.

Companies are increasingly drawing on their user communities to generate promising ideas for new products, which are then marketed as “user-designed” products to the broader consumer market. We demonstrate that nonparticipating, observing consumers prefer to buy from user- rather than designer-driven firms because of an enhanced identification with the firm that has adopted this user-driven philosophy. Three experimental studies validate a newly proposed social identification account underlying this effect. Because consumers are also users, their social identities connect to the user-designers, and they feel empowerment by vicariously being involved in the design process. This formed connection leads to preference for the firm’s products. Importantly, this social identification account also effectively predicts when the effect does not materialize. First, we find that if consumers feel dissimilar to participating users, the effects are attenuated. We demonstrate that this happens when the community differs from consumers along important demographics (i.e., gender) or when consumers are nonexperts in the focal domain (i.e., they feel that they do not belong to the social group of participating users). Second, the effects are attenuated if the user-driven firm is only selectively rather than fully open to participation from all users (observing consumers do not feel socially included). These findings advance the emerging theory on user involvement and offer practical implications for firms interested in pursuing a user-driven philosophy. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mnsc.2014.1999 . This paper was accepted by Pradeep Chintagunta, marketing.

Current branding literature investigates the spillover effects and extension effects due to the introduction of product extensions. However, no study so far has evaluated the aggregate market impact of these effects across different brand development strategies or accounted for the strategic decision to introduce the extension. It is important to examine the above given the significant investments and the high failure rates associated with the introduction of new product extensions. In this study, we develop an analytical framework that derives revenue outcome due to an extension introduction as a function of spillover and extension effects. We empirically estimate the above effects through a Bayesian endogenous switching model that jointly models market shares of the extension and its parent brand along with the strategic decision to introduce the extension and the endogeneity in prices. By using a data set that covers 155 extensions introduced across 20 U.S. geographic markets, we obtain several new generalizable empirical insights. Our results show that spillover effects are higher for brand extensions, whereas line extensions benefit through larger extension effects. We find that vertically differentiating a line extension in terms of increased quality mitigates its negative spillover effects. The addition of a new brand name (i.e., sub-branding) lowers spillover effects for line extensions, whereas it increases the market performance for brand extensions. Our findings provide several strategic implications for manufacturers to successfully introduce and manage product extensions. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mnsc.2014.1900 . This paper was accepted by Pradeep Chintagunta, marketing.

Prominent decision-making theories propose that individuals (should) evaluate alternatives by combining gains and losses in an additive way. Instead, we suggest that individuals seek to maximize the rate of exchange between positive and negative outcomes and thus combine gains and losses in a multiplicative way. Sensitivity to gain-loss ratio provides an alternative account for several existing findings and implies a number of novel predictions. It implies greater sensitivity to losses and risk aversion when expected value is positive, but greater sensitivity to gains and risk seeking when expected value is negative. It also implies more extreme preferences when expected value is positive than when expected value is negative. These predictions are independent of decreasing marginal sensitivity, loss aversion, and probability weighting—three key properties of prospect theory. Five new experiments and reanalyses of two recently published studies support these predictions. This paper was accepted by Yuval Rottenstreich, judgment and decision making.

The consensus wisdom of active mutual fund managers, as reflected in their average over- and underweighting decisions, contains valuable information about future stock returns. Analyzing a comprehensive sample of active U.S. equity funds from 1984 to 2008, we find that stocks heavily overweighted by active funds outperform their underweighted counterparts by more than 7% per year, after adjustments for their loadings on the market, size, value, and momentum factors. This large premium dissipates quickly as the consensus view becomes publicly available. These results are consistent with the notion that informed investing by active mutual funds enhances the informativeness of stock prices. In addition, active mutual funds invest only a small portion of fund assets in high alpha stocks, in accordance with the consensus view that active mutual funds on average fail to outperform passive benchmarks. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mnsc.2013.1847 . This paper was accepted by Brad Barber, finance.

Movie producers and exhibitors make various decisions requiring an understanding of moviegoer's preferences at the local level. Two examples of such decisions are exhibitors' allocation of screens to movies and producers' allocation of advertising across different regions of the country. This study presents a predictive model of local demand for movies with two unique features. First, arguing that consumers' political tendencies have an unutilized predictive power for marketing models, we allow consumers' heterogeneity to depend on their voting tendencies. Second, instead of relying on the commonly used genre classifications to characterize movies, we estimate latent movie attributes. These attributes are not determined a priori by industry professionals but rather reflect consumers' perceptions, as revealed by their moviegoing behavior. Box-office data over five years from 25 counties in the U.S. Midwest provide support for this model. First, consumers' preferences are related to their political tendencies. For example, we find that counties that voted for congressional Republicans prefer movies starring young, Caucasian, female actors over those starring African American, male actors. Second, perceived attributes provide new insights into consumers' preferences. For example, one of these attributes is the movie's degree of seriousness. Finally, and most importantly, the two improvements proposed here have a meaningful impact on forecasting error, decreasing it by 12.6%. This paper was accepted by Pradeep Chintagunta, marketing.

We study customers' adoption and subsequent switching decisions with regard to a menu of three-part tariff plans offered by a commercial bank. Using a rich panel data set covering 70,510 fee-based checking accounts over 30 months, before and after the introduction of the plans, we find that most customers adopt non-cost-minimizing plans, preferring plans with large monthly allowances and high fixed payments. Furthermore, after adoption, customers who exceed their allowances and consequently pay overage fees are more likely to switch to plans with larger allowances than customers who do not experience such fees. Notably, after switching, these overage-paying customers pay higher monthly payments than before. In contrast, switching customers who did not pay overage payments before switching pay less after switching. Our findings, unlike those of previous research on experience-based learning, suggest that the behavior of experienced customers does not converge to the predictions of neoclassical models. We propose that “overage aversion,” which is closely related to loss aversion and mental accounting, is the most plausible explanation for our findings. This paper was accepted by John List, behavioral economics.

This paper proposes and tests a model of antecedents and consequences of group potency in self-managing teams in retail banking. Based on data collected from boundary-spanning service employees organized in 60 teams and their customers, our findings reveal a significant positive impact of group potency on customer-perceived service quality and a negative effect on service profitability. In addition, we find that team consensus regarding group potency positively moderates the effects of group potency, so that for teams with higher levels of potency consensus, the positive impact of group potency on customer-perceived service quality is stronger, whereas the negative impact of group potency on service productivity is weaker. Furthermore, we find significant positive effects of management and interteam support and functional diversity as well as a significant negative effect of team tenure on individual team member beliefs of group potency. Finally, social support consensus moderates the effects of management support, interteam support, and team tenure on group potency, so that the effects of these antecedents are weaker for teams with higher levels of social support consensus. Thus, we conclude that team confidence consensus increases the positive impact of group potency on customer perceptions of service quality and decreases the negative impact on profitability. Thus, team-member perceptual agreement on their team's potency should be stimulated.

In a recent paper, Lippman presents sufficient conditions for Denardo's N-stage contraction in discounted semi-Markov decision processes with unbounded rewards. In this note it is demonstrated that Lippman's conditions may be replaced by weaker conditions which even imply 1-stage contraction. The verification of the conditions of this note is somewhat easier.

This paper investigates the strategic pricing of consumer durable products which can be acquired through either purchase or reproduction (e.g., computer software). As copy piracy results in an opportunity loss, its adverse effect on profits needs to be incorporated in strategic decisions such as pricing. Using a dual diffusion model which parsimoniously describes sales and copying, and employing control theory methodology, optimal price trajectories are derived for the period of monopoly. The results indicate that (a) in absence of any protection, skimming pricing strategies are generally optimal, and (b) copy protection is warranted only when sales diffuse much faster than copying and the protection technology does not significantly raise the marginal production cost.

This paper deals with a one-product production/inventory model, where the production rate can be dynamically adjusted in order to cope with random fluctuations in demand. The inventory level is controlled by using one of two possible production rates where under each production rate the production is continually added to the inventory. The demand process for the product is described by a compound Poisson process. Also, excess demand is lost. In accordance with common practice we consider service measures as the average number of lost-sales occurrences per unit time and the fraction of demand that is lost. For a two-critical-number control rule we derive practically useful approximations for the switch-over level in order to achieve a prespecified service level. Numerical experiments reveal that the approximations are quite accurate.