TIES: Selected Doctoral Theses

TITLE:
“Essays on the Very Invisible College: Global Science and African Participation
– Caroline Viola Fry (2020)
COMMITTEE:
Scott Stern (chair), Pierre Azoulay, Ezra Zuckerman-Sivan
ABSTRACT:
Despite globalization, innovative activities remain concentrated in a handful of high-income
countries. Leveraging knowledge and resources in these locations through ties in the global network
presents opportunities for emerging economies. This dissertation consists of three essays studying the
role of international ties in the development of scientific capacity in sub-Saharan Africa. Each
chapter helps to uncover a different feature of the way in which, and the scope by which,
international ties impact African science, and ultimately facilitate technological catch-up and eco-
nomic growth. Chapter 1 is an introductory chapter, and chapters 2-4 are specific research
applications. Chapter 2 explores the value of international relationships to African scientists
leveraging a unique opportunity afforded to some scientists to develop these relationships: the 2014
Ebola epidemic. Chapter 3 studies the spillover impact of the return home of American trained
scientists to African institutions. Chapter 4 explores a macro-association between foreign knowledge
stocks and African scientific productivity.

TITLE:
“Essays on Innovation and Uncertainty” – Ankur Chavda (2019)
COMMITTEE:
Scott Stern (chair), Pierre Azoulay, Shane Greenstein
ABSTRACT:
In Chapter 1 I study how innovative firms often try out new ideas before fully investing in them
as a kind of experimentation on those ideas. This experimentation generates an early signal of final
outcomes, allowing potentially bad ideas to be terminated before those outcomes are realized. But
not committing to ideas by retaining right to terminate can also be detrimental to outcomes, by for
example attracting lower quality workers or shifting worker effort away from final outcomes towards
passing the experimentation phase. In this paper I explore this tension, asking when does
experimentation improve final outcomes. I test a theoretical model of experimentation against a
dataset of television shows that both enables an estimate of the treatment effect of experimentation
and allows a test for selection bias. I find evidence that experimentation may both handicap worker
recruitment and adversely shift effort. This results in experimentation only improving final outcomes
when it terminates enough bad ideas, otherwise experimentation is detrimental as its benefits are
unable to overcome its downside: the lack of commitment.
In Chapter 2 I consider how entrepreneurs in high growth industries face a unique form of
uncertainty in their search for strategies to execute their ideas: the underlying distribution of potential
outcomes is unknown. This uncertainty creates an opportunity for venture capitalists to extract value
in certain cases by resolving that uncertainty and improving the search prospects for entrepreneurs.
This paper models the optimal search problem faced by entrepreneurs and finds the value
generated by venture capitalists is non-monotonic in the best strategy discovered so far by an
entrepreneur. Our results suggest the rents captured by venture capital may be driven by selection of
a specific kind of entrepreneur: one with a great idea but poor strategy for executing that idea.
1
 In Chapter 3 I investigate the decision to vertically integrate; an important optimization
decision made by firms. However, this decision not only affects the firm itself, it also influences the
firm’s industry as the relationships between firms is changed. This paper is an empirical study of how
vertical integration impacts an industry, specifically the set of new products developed each year:
the direction of innovation. Television shows can either be financed independently of the show’s
broadcast network or partially funded by the show’s broadcast network; this variation in funding
changes the owner of the television show and is therefore a form of vertical integration. Using a
regulatory shock that restricted the networks’ incentives to fund television shows, I find a drop in
vertical integration contemptuous with a shift away from dramas and an overall decrease in the
introduction of new show genre combinations. My results demonstrate how organizational form
affects an industry’s rate and direction of innovation.

TITLE:
“Essays on Economic Sociology of Innovation and Entrepreneurship” – Hye Jun Kim (2019)
COMMITTEE:
Ezra Zuckerman-Sivan (chair), Pierre Azoulay, Matt Marx
ABSTRACT:
This dissertation considers how innovation and entrepreneurship are developed, encouraged,
and evaluated with the theoretical lens of economic sociology. The first chapter investigates who
becomes an entrepreneur among the pool of general consumers. The process by which individuals
become entrepreneurs is often described as a decisive moment of transition, yet it necessarily
involves a series of smaller steps. By collapsing the transition stages of knitting hobbyists’ transition to
producers who sell their original design patterns, the study examines the distinctive characteristics
that affect users’ decision to (a) create new products and (b) commercialize them. In particular, I
show that more experienced, disobedient, and committed knitters tend to make the first transition
and create new products, while knitters who make the second transition and sell their products tend
to be less experienced, disobedient, and committed, compare to the sharing producers who do not
commercialize their products.
The second chapter examines the role of social capital in revealing and encouraging
avocational entrepreneurship. To the question of how social capital benefits innovation and
entrepreneurship, existing literature has provided one dominant answer: the inflow of new information
and knowledge recombination promote innovative ideas. In this study, I suggest a novel insight on
the benefit of social capital on an individual’s transition to avocational entrepreneurs: social networks
provide potential entrepreneurs self-confidence on the promise of their new ideas and encourages
their entry into the market. Using a unique setting in a niche field of knitting, I first show that there are
individuals with great potential to become innovators. Also, using a matched sample of potential
innovators, I show that an individual’s participation in a closely connected local group encourages
her transition to an entrepreneur, especially for those who already have the necessary skills for the
transition. The empirical analysis resonates with qualitative evidence that knitting hobbyists make the
transition to entrepreneurs when encouraged by their friends.
The third chapter (co-authored with Pierre Azoulay and Ezra Zuckerman) considers
commitment-based typecasting among knit designers. We show that “commitment-based
typecasting” has two characteristic features: asymmetry in audience valuation and retrospective
reevaluation. When a novice performer experiences an “identity shock” that suggests that she is
more committed to the audience for one category than another, “betrayed” audience tends to
regard her as having always been less committed to the rival audience/category. We test this theory
in the domain of knitting, where there is a divide between avant-garde knitters and traditional
2
knitters, and we show that when a novice knit designer is first published in the publication associated
with one category, this elicits a retrospective devaluation of her prior work by the audience of the
opposing category.

TITLE:
“Making the Cut: The Rate and Direction of CRISPR Innovation” – Samantha Zyontz (2019)
COMMITTEE:
Scott Stern (chair), Pierre Azoulay, Jeffrey Furman
ABSTRACT:
This dissertation explores, in real time, key institutional factors contributing to the diffusion and
impact of a breakthrough technology from its very first days. The set of studies provide a nuanced
picture of the actors, institutions, technologies, and rules necessary for knowledge managers to make
systematic comparisons among strategies to encourage innovation in emerging industries.
The first chapter examines whether the introduction of a breakthrough technology, the CRISPR
DNA-editing system, affects the trajectory of a scientific field through project selection and new
entry. Using proprietary data from the primary distributor of CRISPR to academic scientists, Addgene,
the study shows that the relative proportion of scientists focusing on editing mammalian cells after the
introduction of CRISPR increased over their counterparts working in bacteria and other eukaryotes.
The shift towards mammalian research may result mostly from entry of new authors.
The second chapter (with Neil Thompson), explores whether characteristics of individual
scientists who experiment with CRISPR differ from those who incorporate that experimentation into a
new project. Using Addgene data we separately observe both groups by matching CRISPR orders to
scientists’ publication histories. We find that some characteristics (e.g., proximity to the discoverers)
do not impact experimentation but do influence the ability to publish, empirically showing that
access to a complex new tool does not automatically translate into the ability to use the tool.
The third chapter builds on the previous two by noting that many new tools require specialized
complementary know-how to be applied effectively and delving into how teams form to acquire
that know-how. Teams in any research domain face the tradeoff of either acquiring this know-how
themselves or working with scarce external tool specialists who also have a choice over domain
teams. CRISPR enables identification of external tool specialists on research teams by exploiting
natural difficulties of applying the tool across disease domains. External tool specialists appear more
often in teams for difficult diseases, especially in subsequent innovations, suggesting that external tool
specialists may be more attracted to complex but influential problems.

TITLE:
“Essays on Learning and Strategy in Research and Development” -- Joshua Krieger (2017)
COMMITTEE:
Pierre Azoulay (chair), Alessandro Bonatti, Fiona Murray, Scott Stern
ABSTRACT:
This dissertation investigates how research organizations learn from and adapt to new
knowledge. In particular, I examine how news about scandals, stigmas and failures influences the
direction of research and development efforts. These negative information shocks force research
organizations to pause, interpret external signals, and apply any lessons to their own project
portfolios. I investigate how these negative information events impact decisions in the settings of
scientific publishing and drug development.
In the first essay, I study the impact of scientific retractions on citation patterns and funding in
the retracted paper's intellectual field. I investigate how the retraction disclosure and affected field's
3
characteristics influence the extent of these spillover effects. The second essay evaluates how
retraction scandals damage individual scientists' reputations. This study shows that the magnitude of
the retraction penalty depends on a scientist's prominence and whether or not the retraction event
involved "misconduct." In the third essay, I analyze how late-stage drug development failures alter
competitor's project continuation decisions. I separate technological learning effects from market
competition effects, and grade decision-making across firms.

TITLE:
“Evaluating Entrepreneurship Programs: Theory and Evidence” -- Daniel Fehder (2016)
COMMITTEE:
Fiona Murray (chair), Scott Stern, Ezra Zuckerman-Sivan, Yael Hochberg
ABSTRACT:
This dissertation consists of three essays studying the impact of a relatively recent type of
entrepreneurship program (startup accelerators) on the performance of firms, regions, and the
selection of early-stage projects in the economy. The first essay (joint work with Yael Hochberg)
explores the impact of startup acceleartors on the level of early-stage entrepreneurial activity in their
region. Recent years have seen the rapid emergence of a new type of program aimed at seeding
startup companies. These programs, often referred to as accelerators, differ from previously known
seed-stage institutions such as incubators and angel groups. While proliferation of such accelerators is
evident, evidence on efficacy and role of these programs is scant. Nonetheless, local governments
and founders of such programs often cite the motivation for their establishment and funding as the
desire to transform their local economies through the establishment of a startup technology cluster in
their region. In this paper, we attempt to assess the impact that such programs can have on the
entrepreneurial ecosystem of the regions in which they are established, by exploring the effects of
accelerators on the availability and provision of seed and early stage venture capital funding in the
local region.
The second essay explores the relationship between a startup's founding region, accelerator
admission and startup performance. Entrepreneurs combine resources from numerous sources as
they build their firms but are constrained by their social and geographic proximity to these resources.
I use this insight as a starting point to explore whether accelerators act as a complement or substitute
for initial location. Using data from MassChallenge, a leading startup accelerator in Boston, I use a
regression discontinuity framework to evaluate both the overall impact of the program on its portfolio
of startups and its heterogeneity based on: 1) the level of entrepreneurship resources in a startup’s
founding region and 2) their ability to access those resources. Startups birthed in neighborhoods with
higher levels of entrepreneurial resources also derive a larger benefit from admission to
MassChallenge. Within the accelerator, startups from richer ecosystems also receive referrals at
higher rates, expanding their social capital relative to entrepreneurs from less rich regions. This finding
suggests that founding regions shape a startup’s performance within the accelerator and that
accelerators change the way in which startup founders are able to access and leverage resources in
their home region.
The third essay explores the actual selection mechanisms inside an accelerator program.
There is a growing awareness that variation in the institutional arrangements used in the selection of
ideas and ventures can have an impact on the types of projects undertaken by innovators. This in
turn shapes not only types of new innovations but also the types of innovators we expect that enter
the economy. To date, research has focused on the composition of selection committees or
differences across quite distinctive evaluation mechanisms (e.g. crowds versus expert committees).
This study hopes to ask a related but unanswered question: Will a fixed set of judges evaluate a fixed
4
set of businesses opportunities differently if they are assigned to different "evaluation regimes".
Specifically, we examine the degree to which status characteristics such as gender and elite
education are critical determinates of project evaluation across two distinct evaluation approaches
– paper-based and committee-based.. We find a strong, positive effect for gender and other
characteristics in a committee-based evaluation scheme where founder characteristics are more
salient to judges. Our findings contribute to a deeper understanding of both the evaluation of early-
stage firms and the role of bias in decision making in settings of considerable uncertainty.