ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

(1)

Isak Trygg Kupersmidt ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

ISBN 978-91-7731-192-8

DOCTORAL DISSERTATION IN ECONOMICS

STOCKHOLM SCHOOL OF ECONOMICS, SWEDEN 2021

Isak Trygg Kupersmidt

ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

This dissertation contains three essays on the theme of risk, privatization, and regulated markets. In the first chapter, the author analyzes the deregulation of the Swedish pharmacy market. Of specific interest is how the market expands and the chains’ decisions on opening new stores. The second chapter is joint work between the author and his supervisor Profes- sor Richard Friberg. They consider a monopolistic competition model with heterogeneous firms and analyze how competition affects the firms’ risk- taking. Using the model, the consequences for the economy of voluntary insurance are compared with mandating it. The third chapter studies the use of private providers in welfare services such as elderly care. The author models the trade-off between possible cost reductions and innovations against the loss of control from introducing competition. The conclusions are then evaluated using data from in-home services for the elderly in Sweden.

The dissertation aims to add to the understanding of the specific markets studied and provide insight into how market outcomes are shaped by regulation.

ISAK TRYGG KUPERSMIDT holds a B.Sc. and M.Sc.

in Mathematics from Stockholm University. His pri- mary field of research is Industrial Organization, with a specific interest in privatization, competition, and market regulation.

(2)

Isak Trygg Kupersmidt ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

ISBN 978-91-7731-192-8

DOCTORAL DISSERTATION IN ECONOMICS

STOCKHOLM SCHOOL OF ECONOMICS, SWEDEN 2021

Isak Trygg Kupersmidt

ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

This dissertation contains three essays on the theme of risk, privatization, and regulated markets. In the first chapter, the author analyzes the deregulation of the Swedish pharmacy market. Of specific interest is how the market expands and the chains’ decisions on opening new stores. The second chapter is joint work between the author and his supervisor Profes- sor Richard Friberg. They consider a monopolistic competition model with heterogeneous firms and analyze how competition affects the firms’ risk- taking. Using the model, the consequences for the economy of voluntary insurance are compared with mandating it. The third chapter studies the use of private providers in welfare services such as elderly care. The author models the trade-off between possible cost reductions and innovations against the loss of control from introducing competition. The conclusions are then evaluated using data from in-home services for the elderly in Sweden.

The dissertation aims to add to the understanding of the specific markets studied and provide insight into how market outcomes are shaped by regulation.

ISAK TRYGG KUPERSMIDT holds a B.Sc. and M.Sc.

in Mathematics from Stockholm University. His pri- mary field of research is Industrial Organization, with a specific interest in privatization, competition, and market regulation.

(3)

Essays on Risk, Privatization and Regulated Markets

Isak Trygg Kupersmidt

Akademisk avhandling

som för avläggande av ekonomie doktorsexamen vid Handelshögskolan i Stockholm

framläggs för offentlig granskning torsdagen den 29 april 2021, kl 13.15, sal Torsten och via Zoom, Handelshögskolan,

Sveavägen 65, Stockholm

(4)

Essays on Risk, Privatization and Regulated Markets

(5)

(6)

Essays on Risk, Privatization and Regulated Markets

Isak Trygg Kupersmidt

(7)

Dissertation for the Degree of Doctor of Philosophy, Ph.D., in Economics

Stockholm School of Economics, 2021

Essays on Risk, Privatization and Regulated Markets

This book was typeset by the author using L^ATEX.

Front cover photo: © Isak Trygg Kupersmidt

Front cover editing: Hugo Trygg Kupersmidt and Caspian Rehbinder Back cover photo: © Alice Hallman

Printed by: BrandFactory, Gothenburg, 2021

Keywords: Industrial Organization, Market Regulation, Privatization, Risk Management, Public Procurement, Pharmacies, Retail, Elderly Care.

(8)

To Ruth and Max

(9)

(10)

Foreword

This volume is the result of a research project carried out at the Department of Economics at the Stockholm School of Economics (SSE).

This volume is submitted as a doctoral thesis at SSE. In keeping with the policies of SSE, the author has been entirely free to conduct and present his research in the manner of his choosing as an expression of his own ideas.

SSE is grateful for the financial support provided by the Jan Wallander and Tom Hedelius Foundation which has made it possible to carry out the project.

Göran Lindqvist David Domeij

Director of Research Professor and Head of the Stockholm School of Economics Department of Economics

Stockholm School of Economics

(11)

(12)

Acknowledgements

First and foremost, I thank my supervisor Richard Friberg for the effort he has devoted to me and my research. I also appreciate him for our joint research, which resulted in the second chapter of this thesis. Throughout my Ph.D. studies, Richard has been invaluable in offering his support and time whenever I have needed it. I could not wish for a better supervisor.

Several people at the Stockholm School of Economics have offered their valuable time and given invaluable comments and suggestions on my research in meetings, over lunches, and during seminars. I am grateful to all staff, students, and faculty I have had the privilege to work with during my time at the school.

I specifically would like to thank Mark Voorneveld for always having an open door.

Regardless of my questions regarded research, teaching, or the Ph.D. program, Mark has always engaged with enthusiasm, wishing to find the best solutions for everyone involved.

I also appreciate having the opportunity to attend and be a teaching assistant at Marks courses in mathematics.

Furthermore, I thank Tore Ellingsen for always being available to discuss questions and problems. Our conversations have been very helpful for my research. I also thank Jörgen Weibull for his interest in discussing ideas with me, regardless of their field of origin. Jörgen has been a great source of inspiration and insight into game theory and economics.

A big thanks go out to my colleagues Atahan Afsar, Albin Erlanson, and Peter Wikman, who has been both helpful and good friends during these years.

Throughout my time at Stockholm School of Economics, the Department of Eco- nomics’ administrative staff has been welcoming and eager to offer their help when needed.

I thank Ritva Kiviharju, Rasa Salkauskaite, Malin Skanelid, and Lyudmila Vafaeva for all the times they helped me during these years. The support they provide to all of us is instrumental for the department and the research conducted here.

My research has been made possible by Jan Wallander and Tom Hedelius Stiftelse, who has provided me with financial support throughout my Ph.D. studies. I am very grateful for their support. I also thank Stiftelsen Louis Fraenckels Stipendiefond for their financial support for my participation in the summer school at the Paris School of Economics.

(13)

viii RISK, PRIVATIZATION AND REGULATED MARKETS

Without the support from my friends and family, I would probably not have made it to the end of this program. I am indebted to all of you for believing in me and rooting for me to succeed. It has meant a lot to me during these years. I am especially indebted to my colleague, office mate, and best friend Gustav Karreskog, with whom I have shared this adventure. Gustav’s contributions to this dissertation are plentiful and too numerous to be listed, so I will only say that Gustav has been instrumental to all my research. Without him, it would have been hard to reach the finish line.

I also thank Caspian Rehbinder, Alice Krön, and Mosa Alasaly for being outstanding friends, always willing to listen and give their views on things.

Some friends have contributed directly to my research. I thank Linus Bergkvist for giving excellent help whenever my skills in mathematics have fallen short and Konrad Hellberg for helping me go through the financial reports of the pharmacies investigated in chapter 1. Furthermore, I am grateful to my friends in economics: Emanuel Welander, Jacob Lundberg, Karl Malmqvist, and Martin Ådahl. I thank you all for discussing ideas with me and offering your insightful thoughts on the matters.

Throughout my life, I have been blessed with a loving and caring family. My mother Judith has always believed in me and encouraged me to find my own path in life while never refraining from speaking her mind. She makes me feel proud of my accomplishments and confident in my abilities when I can not do so myself. My father Robert has inspired my curiosity from my early years and has always been willing to listen to my questions.

My interest in science and society can probably be traced back to the many discussions, on all kinds of topics, we have had over the years. I thank them both for all the time and care they offer me every day. I also thank my wonderful siblings Hugo, Mickel, Rasmus, Karolina, and Johanna, who enriches my life with happiness.

No one wants me to succeed more than my grandmother Ruth. As a teacher, she has always encouraged me to learn new things, always willing to offer her advice while I am at it. Her unwavering enthusiasm for her grandchildren is a never-ending source of inspiration, and I thank her for being there for me whenever I need her.

Finally, my deepest and most sincere gratitude goes out to my fiancé Anna for bright- ening my life with her love and enthusiasm. This dissertation is as much her accomplish- ment as mine. With Anna in my life, everything, even writing a dissertation, becomes a true joy.

Stockholm, March 19, 2021 Isak Trygg Kupersmidt

(14)

Introduction

This dissertation contains three articles with the overarching theme of risk, privatization, and regulated markets. The three chapters approach this broad theme from different perspectives but are unified in their focus on imperfect competition and its relation to market regulation. The first and last articles are single-authored, while the second one is written together with my supervisor Professor Richard Friberg.

I approach these questions using both theoretical models and empirical analysis.

The first chapter on the Swedish pharmacy market is mainly empiric in its approach, the chapter on risk and competition is purely theoretical, and the chapter on cheating in private provisions is theoretical but contains an application to data from the Swedish market for in-home services to the elderly.

Two of the chapters consider markets in the intersection of the public and the private.

These are markets where the government pays for the goods, a public or private actor provides them, and the consumers decide who to buy from. This kind of three-part market is very different from most others as there is no price signal to the consumers, the quantity available to each consumer is fixed, and it is unclear whether it is the consumers or the government who is the actual customer.

The second chapter considers firms in a competitive market, where the government’s role is to regulate the firms’ risk-taking. We evaluate the regulations on insurance against shocks by examining how they affect the market’s long and short-term efficiency. We also examine how competition and risk-taking interact and relate to regulation.

Together, these chapters aim to add to the understanding of competition and market regulation.

* * *

My research is motivated by the realization that there are no truly free markets. Instead, all successful rely on regulation. Laws preventing fraud, banning of certain ingredients, or enforcement of standards and rules are all examples of regulations that can be found in most markets.

If consumers were all-knowing and perfectly rational with infinite liquidity, regulations could, in theory, be unnecessary. Why would anyone, for example, buy poisonous food, a bike that falls apart after a few kilometers, or a drug that does not work? In the

(17)

2 RISK, PRIVATIZATION AND REGULATED MARKETS

simplest of models of rational consumers, regulations can thus seem unnecessary. How- ever, as most information is unknown to most people, and as consumers are bounded in time, liquidity, and cognitive capacity, regulations become necessary to facilitate an efficient market.

For example, it is safe to assume that no-one wants to eat poisonous food, making it much more efficient to ban it than to have each consumer discover what is edible or not. If consumers always need to be wary about if goods are safe, they would be much less inclined to try something new. Thus regulations can facilitate market entry and improve competition and innovation.

Firms can also benefit from regulations. As is known from game theory, the ability to commit can significantly increase cooperation and prosperity in a market. For example, most firms are willing to sacrifice their own option of not fulfilling a contract in return for being able to trust their business partners. Standards and rules can also prevent a race to the bottom and make it possible to compete with higher quality goods and services without being threatened by unserious competitors. Appropriate regulation thus improves the efficiency of the market and can promote both innovation and growth.

As simple as this insight might be, implementation is more complicated. Too much regulation can hamper innovation and instead create barriers to markets by increasing costs and introducing red tape. It can also introduce rent-seeking or introduce perverse incentives. This presents a political choice in most market regulations as different values and goals can collide. For example, strict standards can prevent the emergence of new ways of doing things, while lower standards can increase information asymmetry problems between consumers and producers.

There is no one-size solution for the entire economy. Instead, regardless of the goal, regulations must be tailored to the specific circumstances where they apply. This underlines the importance of understanding how firms compete and consumers behave in different markets, especially those far away from the ideal markets studied in introductory economics.

With this dissertation, I hope to contribute to the understanding of heavily regulated markets, especially at the intersection of the private and public spheres. Such markets are usually not only regulated by politicians but created by them. For example, my first chapter investigates the Swedish pharmacy market where the government, not the pharmacies and the consumers, sets both the prices for prescription drugs and pays for them. The more the state pays the pharmacies for the drugs they hand out, the more money there will be in the market, and the more pharmacies will open. Thus, the market is shaped by the regulations, with a different set of rules resulting in a completely different market.

My research aims to understand the markets I study as they are interesting in themselves and provide insight into how the market outcomes are shaped by how they are regulated.

(18)

INTRODUCTION 3 Cannibalization and Competition

Market Entry Following Sweden’s Pharmacy Deregulation

In the first chapter, I study the 2009 deregulation of the Swedish pharmacy market. The main question is how firms expand in a growing market. Specifically, I analyze how firms decide where to open new stores when there is no competition in price or product differentiation with respect to the physical product. I also estimate how new stores steal market shares from existing ones to see if they compete more with their competitors’

stores or those from the same chain.

The analysis is made possible by a novel data set on the Swedish pharmacy market, where each pharmacy and its revenue is documented every year since the deregulation.

The data have been assembled from open sources and then combined with revenue data from the Swedish eHealth Agency.

The Swedish pharmacy market was deregulated in mid-2009 when the state monopoly was abolished. Private firms were allowed to open new pharmacies, and the state sold three-quarters of their pharmacies to private owners. The deregulation caused a quick increase in pharmacies in Sweden. Between 2009 and 2012, the number of pharmacies increased by 35 percent. This creates a good setting for studying the firms’ decisions on how to expand.

A chain deciding whether to expand in a market must take several factors into account. Firstly, they must consider if there is room for another store in the market at all.

If not, there is no point in entering to start with.

Secondly, the chain must consider the broader consequences of their entry. Suppose a chain has an existing store in the market. In that case, they might steal a substantial amount of customers from themselves, so-calledcannibalization, making it unprofitable on market-level even if the new store might be able to make a small profit. Depending on the market structure, new entries might also cause the market to expand, meaning that the total revenue increases. There are also dynamic effects on the cost side, with having outlets located closer together reducing costs. This is calledeconomics of density.

Thirdly, there are strategic considerations. Competing chains are interacting with each other over time. Opening a store somewhere can thus be used not only as a way of expanding their business but also as either a punishment against a competitor who is too aggressive or to prevent someone else from entering.

My first chapter focuses on the first two aspects, examining how the market size and the chains’ market shares in a local market affect the firms’ decisions.

When studying retail markets, one usually encounters several problems. First of all, most decisions have already been made. Most stores have been at the exact location for decades or more, making the entry decision hard to study and the identification difficult.

This causes researchers to resort to cross-market studies, where they compare similar markets to each other. Another problem is product and price differentiation. Competing

(19)

chains usually sell heterogeneous goods at different prices, making it hard to isolate the forces behind their decisions. One chain may, for example, try to compete with lower prices while someone else is more aggressive in opening new stores. Due to the regulation of the pharmacy market, this is not an issue.

My dataset has several advantages. Firstly, 75 percent of the pharmacies’ revenues come from prescription drugs, whose prices are set by the regulators and are the same across all stores. The goods they sell are physically identical and cannot be bought elsewhere.

Therefore, it is a well-defined market without any meaningful price competition, making it easier to isolate the different effects.

Secondly, the deregulation provides a useful identification: as all entry decisions are made within a few years, they become comparable, allowing me to analyze decisions within markets instead of comparing similar markets to each other. This eliminates the problem of market configuration being path-dependent, i.e., we know that most outlets choose to operate in a market depending on current market conditions, not because of what happened decades ago.

Thirdly, the dataset contains revenue data for each pharmacy, making empirical approximations of revenue redundant. It also becomes possible to measure actual business stealing and cannibalization.

Lastly, Sweden has excellent demographic and spatial data to match with the market data. Together this makes it possible to isolate the issue of business stealing, market cannibalization, and economics of density to see how these affect entry thresholds and entry decisions.

My analysis provides clear evidence of the chains wanting to open new stores in markets where they are incumbent. At the same time, chains are less likely to open stores in markets where they already have a significant market share. Previous studies on retail competition have also documented a higher number of stores per inhabitant in smaller markets, attributed to more intensive price competition in larger markets. I find no such effect, which is expected as there is no price competition between the pharmacies.

Using revenue data at the pharmacy level, I also estimate cannibalization and business stealing, finding that new stores steal twice as much revenue from their competitors than from their own stores.

The findings provide additional insights into pharmaceutical markets and add another layer of support to previous conclusions about retail competition.

Does stronger competition drive out precautionary behavior?

with Richard Friberg

The second chapter is joint work with my supervisor, Professor Richard Friberg. In our paper, we develop a theoretical model of how firms react to the risk of a large shock hitting the economy.

(20)

INTRODUCTION 5 The central idea is that firms face a trade-off between cost efficiency and the robust- ness of their operations. For example, firms that keep larger stocks of goods and parts are less vulnerable to disturbances in their supply chain but will have higher production costs.

This makes more stable companies less competitive, exposing them to the risk of being driven out of business by more reckless competitors. On the other hand, if an external shock hits the market, they will thrive while their competitors go out of business. We model this trade-off as a discrete choice between the two alternatives of having a higher marginal cost for producing goods but not being vulnerable to external shocks or having lower costs and risking going out of business if something unexpected occurs.

We build on an existing model by Melitz and Ottaviano (2008), where firms decide to enter a market and then discover their marginal cost of producing their unique good. This creates a spectrum of firms with different productivity that sells slightly different goods, resulting in the more productive firms getting larger market shares than the less productive ones. The firms with too high costs will choose to leave the market, resulting in an equilibrium where the number of firms in the market is dependent on the circumstances.

We see that if firms are free to choose whether to insure themselves, only the most efficient firms will do so as they are the most profitable, and thus those with most to lose if they go out of business. The least productive firms already have such high costs that they would not be able to profitably operate if they had to pay for insurance. This finding is in line with empiric evidence but at odds with standard motivation for risk management.

If insurance instead is mandatory, e.g., mandated by the state, we find that competition will be tougher with lower marks-ups and more firms in the market. On the other hand, prices can be higher as insurance increases the firms’ costs. More specifically, we find that mandatory insurance is preferable for the consumers when the risk of a shock is high, but insurance costs are low.

We also find that competition does drive risk-taking, but only for the least productive firms. The availability of insurance introduces more competition as the most profitable firms’ expected profit goes up, causing more firms to enter the market. The increased competition reduces all firms’ margins and makes the least productive ones unable to pay for insurance, forcing them to take risks to make a profit. Furthermore, we find that in markets with higher competition, fewer firms will insure themselves.

Looking at the consequences of shocks occurring, we find that if a shock hits the economy, mandatory insurances mean that all firms survive. In the short run, this is good for consumers and firms, as the variety in the market does not change, and no firms go out of business.

In the long run, however, mandatory insurance can be bad for the economy. The firms that do not insure themselves if insurance is voluntary are the least productive ones.

If they go out of business, new firms can enter the market. Some of which will be more productive, meaning that in the long run, shocks that remove the least productive firms

(21)

from the market can be good for the economy as it increases productivity and thus reduces prices for the consumers.

In conclusion, we find that voluntary insurance is optimal when new firms are quick to emerge, and the costs of starting them are not too high. In comparison, mandatory insurance is preferable when the economy can sustain the short-term effects of multiple firms going out of business. This will both keep costs down and increase productivity over time. For example, it might be a good idea to force banks to insure themselves against shocks but let restaurants compete as they like. The result of such policy would be more innovation among restaurants than banks, but with more bankruptcies and fluctuations in the supply of restaurants as a consequence.

Cheating in Private Provision of Services

A Theoretical Model with an Application to In-Home Services for the Elderly The last chapter of my dissertation builds on the idea that availability of governments services should differ depending on how easy it is to verify that they are provided correctly.

For example, it is easy to diagnose a broken leg and prove that correct treatment has been given. Thus, we should expect care for broken legs to be fairly accessible. Meanwhile, mental illnesses are harder to diagnose and verify, and the appropriate treatment is not obvious, making mental care harder to get for those in need of it.

This would explain why there can appear to be underinvestment in certain areas by governments. Also, this effect should be even more prominent in systems where private firms are tasked with diagnosing problems and offering the appropriate remedy. It would also imply that the use of private providers is less suitable for harder-to-diagnose problems.

Twelve percent of GDP in the OECD countries is spent on goods and services bought through public procurement by the governments (OECD, 2017). At the core of this is a "make-or-buy" choice, where the governments must make a trade-off between buying goods on the market or having them produced by government employees. In- house production can increase the governments’ control over the production but limits competition. On the other hand, private entrepreneurs can introduce competition resulting in cost and quality innovation but reduce the government’s control over what is actually provided.

A private company manager has very different incentives than a government em- ployee, especially when it comes to reducing costs. Under the right circumstances, one could see public procurement resulting in competition in both quality and price, leading to cheaper and better services for the public. At the same time, the interests of private managers do, in general, not coincide with the public interest, leading to concerns about lower quality in non-contractable aspects. This is especially important for services such as health care, where outcomes and the providers’ performance can be hard to verify.

(22)

INTRODUCTION 7 In this third chapter, I examine this make-or-buy decision by modeling how the locel government’s demand for goods and services is affected by the possibility for private providers of these goods to cheat or deviate from contracts. I develop a theoretical model and test the results on data from in-home services for the elderly in Sweden. I also examine how incentives to cut costs, both by innovation and cheating, affect the local government’s optimal choice and their demand for these goods. The model builds on the model presented in Hart et al. (1997) and extends it to look at how the possibility of non-contractable cheating affects the local government’s demand (and thus the supply of government services to the public).

The model provides insights into the choice between public and private providers of goods. I find that private providers are optimal when the possibility for cost innovation is good and when it is easy to detect cheating.

The model also shows that the possibility of cheating leads to a lower government demand when using private providers. Even when privatization is optimal and cheaper, it might lower the demand for services due to the risk of cheating. Privatizations can thus reduce public spending and increase the social surplus, but at thecost of reduced quality per unit and, in some cases, lower demand from the government (causing a lower supply of public goods).

The empirical part of the chapter uses data on in-home services for the elderly provided by the Swedish municipalities. The municipalities decide whether they want to provide the service using publicly employed personnel or allow private providers. In either case, the cost is covered by the municipality.

In general, I find that lower costs coincide with a higher demand by the local government (more hours provided to the elderly) and that a larger share of private providers coincides with lower costs per hour. However, more privatization does not correlate with government demand, suggesting that privatization counteracts the effect on demand from lower costs. In other words, private providers coincide with lower costs and fewer hours of care provided to the elderly.

Thus, the empirical analysis results are in line with the model’s predictions — privatizations correlate with lower costs and appear to limit the municipalities’ demand for the services.

* * *

This dissertation aims at adding to the understanding of regulated markets from both an empirical and theoretical point of view. If I were to draw some general conclusion from my research, it is this that every market requires their own regulation and trade-off between public and private interests. There is no such thing asprivatization and regulation, but numerous different solutions with more that sets them apart than unifies them. Any policy must acknowledge this to be successful.

(23)

ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

Isak Trygg Kupersmidt

ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

Isak Trygg Kupersmidt

ESSAYS ON RISK, PRIVATIZATION AND REGULATED MARKETS

Essays on Risk, Privatization and Regulated Markets

Isak Trygg Kupersmidt

Akademisk avhandling

som för avläggande av ekonomie doktorsexamen vid Handelshögskolan i Stockholm

framläggs för offentlig granskning torsdagen den 29 april 2021, kl 13.15, sal Torsten och via Zoom, Handelshögskolan,

Sveavägen 65, Stockholm

Essays on Risk, Privatization and Regulated Markets

Isak Trygg Kupersmidt

Foreword

Acknowledgements

Contents

Introduction