Entrepreneurial Decision and the Strange Behavioural Economics of Elon Musk

PETER J PHILLIPS
Faculty of Business, Education, Law and Arts
University of Southern Queensland
Toowoomba, Australia

GABRIELA POHL
Faculty of Business, Education, Law and Arts
University of Southern Queensland
Toowoomba, Australia

Abstract. We look at several of Elon Musk’s business ventures, forays, and interests: (1) X.com; (2) SpaceX; (3) Tesla; (4)
cryptocurrency; and (5) Twitter. We discuss Musk’s decision-making and see whether behavioural economics, especially
prospect theory, can help us to better understand his motivations and choices and the ways in which he proceeds in the
face of true uncertainty. Recent efforts directed towards connecting decision theory with theories of entrepreneurial
behaviour, including those developed by the Austrian School, are examined in the context of Elon Musk’s decisions.

Keywords. Elon Musk, Behavioural Economics, X.com, Tesla, SpaceX, Cryptocurrency, Twitter, Entrepreneur, Austrian
School

Introduction

Writers had the blank page; Robert Evans had his dial tone. All his imagining—his multilayered consideration
of scripts and how to get them into movies—began here, on the phone, with slightly more than nothing, just
seven digits and a hunch. What about Faye Dunaway for Chinatown? What about Jane Fonda? Would she
come for dinner this week? He wanted to talk to her. He wanted to hear her ideas…. These invitations were
stepping-stones Evans would place across dry river beds. Then he would step back and survey his progress,
and ask, Will those get us to a movie? What else do we need?1

Behavioural economics is a list of plausible and comprehensible, though not immediately obvious things that
have been learned about how humans make decisions. We can use items from this list together with what we
know from our own experience to gain a better understanding of how people make decisions and why they
do what they do. This has worked well up till now. It is the reason behavioural economics has been accepted
as a worthwhile addition to knowledge. The possibility remains, though, that someone’s decisions might defy
interpretation and yet be so successful, so compelling, as to cause us to rework our list, or to put a few
asterisks alongside some of our items. Or, perhaps, we will be compelled to complement our list with insights
generated by other parts of economic theory.
Purely rational decision-makers are never buffeted hither and thither by either the context or their
own feelings or predispositions. But people are not always so clear-minded or clear-sighted. Their mistakes
are repeated, and their decisions shaped by influencers, nudgers, and mavericks. These are the people who
set the social stage for decision-making. While all decision-makers who wish to survive for any length of time
must adapt to innovations, more than usually prominent disruptors may shake the very foundations of the
problem space within which decision-makers had thought they were operating. It is especially important,

 Address correspondence to Peter J Phillips ([email protected]) and Gabriela Pohl ([email protected]) at the
University of Southern Queensland, Toowoomba, Queensland, Australia, 4350.
1 Wasson (2020, p.146). Robert Evans became head of Paramount soon after Gulf + Western’s purchase of the studio and

oversaw its revival as a prominent player in the American New Wave (or New Hollywood) that blossomed in the 1970s.

Electronic copy available at: https://ssrn.com/abstract=4253679

therefore, to understand the entrepreneurial decision-making of these types of individuals, even more so than
the ‘typical’ entrepreneur, if there is such a thing.
This brings us to Elon Musk. He has certainly played the role of eccentric disruptor to perfection. And
so, we look at several of Musk’s business ventures, forays, and interests: (1) X.com; (2) SpaceX; (3) Tesla; (4)
cryptocurrency; and (5) Twitter. Doing so can help us organise and make sense of what is a somewhat unusual
collection of decisions. We can also see how behavioural economics performs in assisting us with this task.
Most importantly, we can see how flexible the frameworks of behavioural economics are at absorbing
entrepreneurial activity of a most interesting kind and whether there is anything to be gained by
complementing the insights that might emerge from a behavioural economics analysis with the existing theory
of entrepreneurial decision-making. This also leads us to explore the connections between decision theory,
both orthodox and behavioural, and the theory of the entrepreneur, especially the entrepreneur’s navigation
of pure uncertainty.

X.com
The first online or “home” banking services were launched in 1980 (Cronin 1998). In 1999, 5 percent of
American households did their banking online, well below the enthusiastic estimates of the mid-1990s. In fact,
the journal for the American Banker’s Association (ABA) carried an article in 2000 entitled, “Online Banking
Yet to Deliver”, in which the author argues that online banking “…may never be more than a tool for unique
groups” (Bielski 2000). In the United States, between 1995 and 2003 electronic banking use increased from
around 150,000 users to 3.2 million. This was just the beginning. By 2004, 53 million people or 44% of Internet
users and one-quarter of all adults were using online banking (Servon & Kaestner 2008, p.276). By 2020, 14.2
million Americans said that a digital bank rather than a traditional “bricks and mortar” bank with an online
banking service was their primary bank (Shevlin 2020). By 2021, 76 percent of Americans said they banked
using a mobile app (Strohm 2021) and there were almost 200 million Americans using digital banking services
(Business Insider 2021).
In 1999, when Elon Musk, Harris Fricker, Christopher Payne, and Ed Ho founded X.com, there had been
limited uptake of online banking services, and industry insiders were sceptical that the idea had any
appreciable upside, even for established financial institutions with an existing customer base. Establishing a
purely online bank seemed especially kooky. Soon after its founding, Nath, Schrick & Parzinger (2001, p.24)
had this to say about X.com:

X.com Bank is the first bank operating in a “Silicon Valley culture” where seizing customers takes
precedence over making a profit during a company’s growing years. The bank is in the process of buying
the brick-and-mortar First Western National Bank of Colorado which currently performs its banking
services. X.com, a start-up bank with no foundation in the banking industry, represents the kind of threat
brick-and-mortar banks are facing from Internet only e-banks. With few barriers standing in the way, anyone
can enter the banking market and offer customers innovative banking solutions, without the overhead of a
large infrastructure (Lewis, 2000).

X.com was not Musk’s first business venture. In 1995, Musk and his brother had set up what became Zip2, an
online “city guide” service, that also helped connect customers and advertisers. In 1999, Zip2 company was
sold to Compaq Computer for more than $300 million. Musk netted around $20 million and used about half
of it to start X.com. The company got off to a strong start, allowing users to do things like send money to
another person through email addresses. X.com merged with Confinity in early 2000. Confinity’s flagship
product was PayPal. While the name X.com was initially retained, the company was later renamed PayPal, and

Electronic copy available at: https://ssrn.com/abstract=4253679

went public in 2002 (under the leadership of Peter Thiel). PayPal was acquired by eBay shortly thereafter for
$1.5 billion. Musk’s share of the PayPal sale to eBay was more than $175 million. PayPal is now a household
name but even before X.com merged with Confinity, it encompassed online banking, money transfers, and
glimpses of FinTech at a time when the industry itself had doubts about the future of internet banking and
relatively few Americans had adopted it.

SpaceX
Musk likes the letter “X”. SpaceX started with an idea about growing plants on Mars. In pursuit of this idea,
Musk tried to purchase some second-hand Dnepr rockets (1999 model) from Russia. He couldn’t strike a deal,
so he decided to build the rockets himself. Beneath this veneer of whimsy, Musk had identified an opportunity
for entrepreneurial profit. The opportunity was fairly “obvious”. To send a 550-pound payload into orbit cost
upwards of $15 million ($30,000 per pound). If that cost could be reduced, SpaceX could be the hottest thing
since the glory days of Texas Air. Musk believed that he could indeed get that cost down to around $3 million
($5,500 per pound). Established companies such as Boeing and Lockheed Martin would not be able to
compete, at least not for a while. And the potential demand from companies and governments seeking to
launch satellites into space at such a bargain price might be only a fraction of opportunities available to key
players in a new space race.
It took time and more money than Musk had originally thought but in 2008 SpaceX became the first
private company to launch a rocket into orbit. More firsts followed, including becoming the first private
company to successfully return rockets to Earth intact and orchestrating an upright landing after returning
from an orbital flight. Reusing materials, of course, is a key part of the cost savings that underpin SpaceX’s
business model. Following these successes, SpaceX secured contracts with NASA to deliver equipment and
crew to the International Space Station (ISS). It also has defence-related contracts. However, the bulk of its
money seems—being a private company, the details are opaque—to come from satellite launch services, from
which it generates significant revenue. SpaceX charges approximately $70 million per launch, which at a price
of $1,200 per pound is a fraction of the $30,000 per pound that was previously the going rate and below
Musk’s original price point. $70 million per launch is very cheap when it is considered that NASA’s shuttle
program cost around $1.5 billion per flight (Chow 2022).
Alongside this sits the fast-growing Starlink satellite internet business. By mid-2022, there were more
than 2,500 satellites supporting the Starlink business, the aim of which is to provide a satellite internet service
to everywhere on the planet. At the same time (mid-2022), there were more than 400,000 subscribers to the
internet service across 36 countries. That represents rapid growth over the March 2022 subscriber base of
250,000 (Sheetz 2022a). The 2022 subscription cost was $110 per month plus $599 in set-up costs, for the
satellite dish etc. (Sheetz 2022a). This business could be spun-off from SpaceX and taken public on its own.
However, Musk suggests that this might not happen until 2025, at the earliest (Sheetz 2022b). If subscriber
growth continues, it could be a spectacular stock market debut for Starlink. With 400,000 current subscribers
at $110 per month, revenue is already (in mid-2022) $520,000,000 per year.
The bigger question concerns the value of SpaceX as a whole. Valuing privately held companies is
tricky but there are some indicators that reveal just how much people think SpaceX is worth. Unfortunately,
smaller investors have few options for investing in SpaceX beyond purchasing shares in Alphabet (parent
company of Google) which, together with the funds manager Fidelity, invested $1 billion in 2015 for a 10
percent stake in SpaceX. Logically but very roughly speaking, if 10 percent was worth a billion dollars, then

Electronic copy available at: https://ssrn.com/abstract=4253679

the whole company must have been worth around $10 billion in 2015. Further valuation glimpses come several
years later. In late 2018, SpaceX raised $500 million dollars from existing investors through the sale of
approximately 2.5 million new shares at $186 per share. This price apparently valued the company at $30
billion (Winkler 2018). In August 2020, the company raised another $2 billion at an implied whole-of-company
valuation of $45 billion. Then, in early 2021, the company raised a further $850 million, this time at a price of
$419.99 per share and an implied whole-of-company valuation of more than $70 billion (Sheetz 2021). By mid-
2022, some commentators pegged the value of SpaceX at $125 billion (Tan & Roof 2022). Elon Musk owns
around 50 percent of SpaceX (Root 2021).

Tesla
By the early 2000s, Musk had amassed a fortune of around $200 million. In 2004, he invested $6.5 million to
become the largest shareholder in Tesla, which had been incorporated by Martin Eberhard and Marc
Tarpenning the year before. Musk has been Chairman since 2004 and CEO since 2008. Five years later,
Tesla had delivered less than 150 cars and Musk had personally contributed an additional $60 million of
capital. Demand remained subdued, including for the first generation “Roadster” model which sold 2,500 units
from 2008 until being discontinued in 2012. As the production figure just quoted suggested, there was little
initial or even medium-term excitement about Tesla’s prospects. The stock price chart in Figure 1, covering
the period June 2010 to June 2022, says a thousand words.

Figure 1 Tesla Inc. Stock Price June 2010 to June 2022

1400

1200

1000
Stock Price ($US)

800

600

400

200

By 2010, both Musk and Tesla were facing serious cash flow problems. In court filings pertaining to his divorce,
Musk declared that he had run out of cash in 2009 (Thomas 2010). At the same time, Tesla’s cash burn rates
were at worryingly high levels. The situation eased somewhat with a $465 million loan from the U.S.
Department of Energy in 2009. Indicating the extent of the company’s cash burn, it used up 10 percent of the
Department of Energy Loan within twelve months (Andrejczak 2010). In June 2010, the company went public
at $17 a share, raising more than $200 million in capital.
Early shareholders were not rewarded with immediate capital gains but innovation in terms of new
models and new products (e.g., batteries) proceeded at pace. In 2016, Tesla manufactured and shipped

Electronic copy available at: https://ssrn.com/abstract=4253679

75,000 vehicles. Then 100,000 the following year. Then 245,000 in 2018, 367,000 in 2019, and 500,000 in
2020. In 2021, Tesla shipped 935,000 electric vehicles (comparable with Toyota’s total, but far exceeding
Toyota’s electric vehicle sales). Battery power is not Tesla’s only innovation. The company’s experiments with
varying degrees of “autopilot” capability are well known. These features, leveraging developments in artificial
intelligence, could potentially open significant new markets for Tesla (e.g., RoboTaxis). On the back of strong
sales, Tesla’s 2021 profit soared to more than $5 billion (Boudette 2022). The shareholders, new and old, were
rewarded with stellar growth, far above that experienced by shareholders in the Ford Motor Company, whose
stock price performance is charted in Figure 2. In fact, Tesla had become by far the most valuable automaker
in the world, bigger than many other manufacturers combined.
In mid-2022, despite significant price corrections across the tech sector, Tesla’s market
capitalisation of $800 billion was almost four times that of Toyota’s and more than 10 times that of Ford’s.
Some commentators point out, however, that Tesla’s price-to-earnings ratio of more than 100, compared to
a price-to-earnings ratio of just 10 for Toyota and a miniscule (by comparison) 4.88 for Ford, indicates that
Tesla is dramatically overvalued. Of course, who is right and who is wrong will be decided in time. Elon Musk
owns around 17 percent of Tesla or around 170 million shares (Frank 2022).

Figure 2 Ford Motor Co. Stock Price June 1972 to June 2022

40

35

30
Stock Price ($US)

25

20

15

10

0
1/06/2000
1/06/1984
1/06/1972
1/06/1974
1/06/1976

1/06/1980
1/06/1978

1/06/1982

1/06/1986
1/06/1988
1/06/1990
1/06/1992

1/06/1998

1/06/2004
1/06/1994
1/06/1996

1/06/2002

1/06/2006
1/06/2008

1/06/2020
1/06/2022
1/06/2010
1/06/2012
1/06/2014
1/06/2016
1/06/2018

Cryptocurrency & Twitter

Elon Musk is known as a “crypto bro” and his tweets and commentary have sometimes impacted crypto
markets significantly (Molla 2021). However, he doesn’t appear to have been an especially early adopter of
cryptocurrency or its underlying and related technologies. His first public comments of note about
cryptocurrency came in 2019, in an episode of the Ark Invest Podcast. Among other things, Musk said, “Paper
money is going away, and crypto is a far better way to transfer value than pieces of paper.” Through the later
part of 2019 and into 2020, Musk waded into the crypto debate with typical swashbuckling aplomb. When
JK Rowling tweeted that she didn’t understand the risks involved with crypto, Musk responded by tweeting
that loose monetary policy from central banks made crypto look stable in comparison to fiat currencies. He
also played a role in launching Dogecoin to prominence, implying that it would come to dominate the financial
system. Often, Musk’s communications about crypto were vague, mysterious, tongue-in-cheek, mischievous,

Electronic copy available at: https://ssrn.com/abstract=4253679

or otherwise designed (or not) to stir the pot. He has also shifted or abruptly changed course. Most famously,
declaring in early 2021 that Tesla would accept bitcoin and then reversing the decision a few months later.
And, just in case one happened to be thinking that this meant that he had changed his mind on crypto
completely, Tesla continued to hold around $250 million in bitcoin on its balance sheet in mid-2022, after
selling down its holdings during July (Bursztynsky 2022, BBC News 2022).
Much of what Musk says about cryptocurrency (and everything else) is communicated via Twitter, in
a mixture of memes as well as text. So, there was a certain symmetry when Musk announced in early 2022
that he had purchased a sizeable stake in Twitter. Shortly thereafter, he offered to take the company private
at a price of $54.20 per share in a deal worth $44 billion. Between April and October 2022, in a bargaining
game that often bubbled over into (no surprise) a public exchange of tweets, the deal remained up in the air.
As the Twitter share price hovered below the $54.20 offer price, Musk and Twitter representatives disputed,
among other things, the number of “real” Twitter accounts. This is a critical piece of data because too many
fake or inactive accounts significantly reduces the reach of advertising and, hence, advertising revenue.
Interestingly, leaving aside Musk’s claim that up to one-in-five Twitter accounts may be “bots” and Twitter’s
counterclaim that it’s less than one-in-twenty, there are other serious issues with usership. For example, while
there are 1.3 billion Twitter accounts, only around 15 percent of them are active daily. And many users have
more than one account. Furthermore, in Australia, less than 3 percent of 16-to-64-year-olds rank Twitter as
their favourite platform (Tong 2022). In early July, probably not surprisingly, Musk announced that he had
withdrawn his offer. Some commentators have interpreted this as the end of the road for the deal. Others
think it is just another negotiating tactic and that Musk will be back with another (lower) offer.
Lost in the maelstrom of the Twitter deal, including chatter about free speech and censorship, is the
relevance of Musk’s proposed Twitter takeover for cryptocurrency. Bloomberg analysts Davies & Eltzroth
(2022), however, noticed the connection. As we have mentioned, Musk uses Twitter to communicate his
thoughts about cryptocurrency, and his tweets can move markets. The title of the Bloomberg piece is: Crypto
Boost May be Focus for Elon Musk as He Acquires Twitter. To support their argument, Davies & Eltzroth
(2022) present a snapshot of the crypto-world’s reaction to the Musk Twitter deal. Positive tweets (of course)
came almost immediately from Bitfinex and Coinbase. The latter, one of the most prominent cryptocurrency
exchanges, quickly elaborated with a deeper explanation published on the company’s website (Coinbase
2022). They listed the possible expansion of several recent crypto-related Twitter integrations including
USDC stablecoin payments, NFT profile photo verification, and BTC (bitcoin) tips for creators as reasons why
Musk’s takeover was a positive sign for the crypto world in general. More deeply, there is a sense of shared
philosophy. The possibility that Musk could “decentralise” the Twitter algorithms and make Twitter a Web3
hub aligns with the core tenets of the decentralised finance movement. A Musk-run Twitter could give crypto
a boost and, in turn, crypto could be a boost for Twitter. If Musk sees that angle, and there is reason to expect
that he does, then the Twitter deal is more than users and advertising revenue. And much more than a ‘free
speech’ initiative.

Structuring the Entrepreneur’s Problem

The economics of information (decision theory) is concerned with the rankings of alternatives under
conditions of risk and uncertainty. It is about a decision-maker confronted by a problem space described by
alternatives (things that can be chosen), possible outcomes (denoted by 𝑥) and probabilities (𝑝’s). The problem
space is closed, though it may not be completely ‘populated’ at the beginning of the decision-making process.

Electronic copy available at: https://ssrn.com/abstract=4253679

The alternatives are the access points to the outcomes (𝑥’s). A decision-maker must make a choice before
he can experience one of the outcomes. The choice depends on an evaluation of the alternatives based on
the expected outcomes. Although the decision-maker does not possess all the information about alternatives,
𝑥’s and 𝑝’s, he or she can possess it if enough time and money is spent on ‘search’. During the search phase,
the decision-maker receives information signals (denoted by 𝑦) and uses the 𝑦’s (from a signal space) to
uncover 𝑥’s in the problem space. This process is depicted in Figure 3. Both spaces are bordered and can, in
principle, be completely populated.

Figure 3 The Closed but Fillable Problem Space

Within this framework, the entrepreneur must be someone who ‘fills’ the empty areas of the problem space
better or faster (or both) than everyone else and then uses that advantage to select the best alternative. While
the entrepreneur is better and faster than others, everyone would eventually have spotted the best alternative
because time alone will sooner or later resolve the uncertainty. No human action is necessary to achieve that
resolution. Kirzner (1997, p.70) contrasts this model of the decision problem, which underpins neoclassical
microeconomics, with the open, uncertain, and unfillable problem space of Austrian economics that frames
the Austrian School’s ongoing struggle to develop a deeper theory of entrepreneurial action:

Austrian theory thus diverges sharply from the notion of the individual decision that constitutes the analytical
building block of neoclassical microtheory. For neoclassical microtheory each decision, whether made by
consumer, firm, or resource owner, is made within a definitely known framework made up of a given objective
function, a given set of resource constraints, and a given set of technologically or economically feasible
ways of transforming resources into desired objectives. Uncertainty, while of course recognized as
surrounding each decision, expresses itself in the form of known probability distributions relating to the
given elements of this known framework.

In place of a mechanical search and decision process within a closed and fillable problem space, the Austrian
school has traditionally recognised the open nature of most important real-world problems and explores
human action in the face of this uncertainty. The entrepreneur’s actions are necessary to resolve the
uncertainty. It won’t be resolved simply by the passage of time. The Austrian criticism of the orthodox

Electronic copy available at: https://ssrn.com/abstract=4253679

treatment of decision and, by extension, entrepreneurial decision-making, has contributed to our
understanding of the true nature of the challenge facing the entrepreneur. Kirzner (1997, p.62) says:

What stamps the entrepreneurial discovery approach as Austrian is not these criticisms themselves, but
rather the specific positive elements of the approach. These positive elements focus on the role of
knowledge and discovery.

The knowledge Kirzner refers to is ‘un-thought-of knowledge’ or ‘unknown ignorance’. Discovery is the
process by which unknown ignorance, which might be described as ignorance of un-thought-of knowledge,
is resolved. From this perspective, ‘alertness’ guides the entrepreneur’s actions as he or she navigates through
uncertainty vis-à-vis ‘search’ in neoclassical theory which allows the entrepreneur to identify the location of
each steppingstone through the problem space before he or she even takes a step. The difference is deeper
than that, of course. The Austrian School’s entrepreneur faces true uncertainty, a situation where the
alternatives, 𝑥’s and 𝑝’s are not completely knowable, even in principle. And it is the entrepreneur’s action that
helps resolve uncertainty. There seems to be an impasse, which might be putting it mildly. ‘Alertness’ and
‘search’ sit atop two fundamentally different interpretations of the nature of the problem space that
entrepreneurs and other decision-makers confront and somehow traverse.
Within Austrian economics there are, roughly speaking, two dominant theoretical standpoints on this
issue. On the one hand, we have those aligned with Kirzner (1973, 1985, 1997) or, more generally, those who
see the essence of entrepreneurial decision-making encompassed by concepts such as discovery and
alertness. On the other hand, we have those who see entrepreneurial decision-making as a species of
judgement under uncertainty and, therefore, possibly integrable with the economics of information (see Klein
& Bylund (2014) for a review). The latter approach involves taking Austrian analysis and directing it towards
the uncertainty problem that orthodox theorists thought could be resolved (or dismissed) by search theory,
in the process connecting the more familiar concepts of decision theory with the less well-known concepts
of Austrian theory. Essentially, it would represent an attempt to answer the question as to how or through
what human action the problem space is made manageable enough to permit a decision to be made that is
not simply a leap of faith into the unknown and not merely a matter of information gathering.
Work is slowly progressing, but the general direction of research and findings so far are reflected in
work by authors such as Foss & Klein (2012) and Packard, Clark & Klein (2017). They discuss two different
types of reasoning that entrepreneurs might use to resolve true (Knightian) uncertainty. The first is causal
reasoning. The second is effectual reasoning.
As we mentioned, standard decision theory covers some sort of search for alternatives, 𝑥’s and 𝑝’s
and, following the population of the problem space, applies an ordering or ranking procedure (e.g., expected
utility). The alternatives are the doorways through which the outcomes are accessed. The idea is to find and
go through the best doorway. Usually, a goal or problem has been pre-defined. For example, find the best
investment for a moderate-risk portfolio. The process begins by identifying investments (alternatives) and
their possible payoffs. The process ends with the ranking of the investments and the selection of the best-
ranked alternative. To the extent that there ever was true uncertainty in this framework, it has been reduced
to risk. Upon completion of the search, all the outcomes and probabilities are known. There are no grey areas
in the problem space. There are no surprises (i.e., an outcome that no-one expected).
Using either causal or effectual reasoning, the entrepreneur approaches the structure of decision
from a different angle. If a problem has been defined, the decision-maker who follows a causal reasoning

Electronic copy available at: https://ssrn.com/abstract=4253679

pathway does not proceed by populating a problem space with alternatives, 𝑥’s and 𝑝’s. Rather, the decision-
maker works on listing all the outcomes that might be associated with solving the problem. The goal is to
determine whether there are any valuable or worthwhile outcomes to be experienced by solving it. If not, the
decision-maker moves on to the next problem. If so, the decision-maker works on finding an alternative that
solves the problem and opens the doorway to one of the valuable outcomes that has been identified.
Sometimes, a problem may not be defined. In this case, the decision-maker might follow an effectual reasoning
pathway where he or she works on listing a set of alternatives rather than outcomes. Once the uses to which
the available resources can be put have been listed, the decision-maker next determines whether there are
any valuable outcomes that might be reached by using one of the alternatives as a doorway through to them.
These two processes might be illustrated as shown in Figure 4. The question is whether there are any parts
of these different reasoning structures that can be clarified by behavioural economics.

Figure 4 The Inverse S-Shaped Probability Weighting Function

Behavioural economics was born within the structure of orthodox decision theory and is applied within that
structure with the purpose of explaining why people fail to follow the prescriptions of expected utility theory.
That is, why people make systematic mistakes when assessing alternatives, 𝑥’s and 𝑝’s. Prospect theory forms
the theoretical core of behavioural economics because it is a formal generalisation of expected utility theory.
Without it, behavioural economics would be a useful but somewhat loose collection of findings. For example,
we can say that Herbert Simon’s satisficing concept is an explanation for why people stop looking for
alternatives and decide there and then. Or we could say that Richard Thaler’s mental accounts concept is an
explanation for why people fail to think in terms of their overall wealth (or portfolio). Contrast such
explanations with prospect theory, which can be applied to a set of risky prospects to show how several
features of the human decision-making process disrupt the ordering of those alternatives and drag it away
from the ordering prescribed by expected utility theory. While both theories require a set of alternatives, 𝑥’s
and 𝑝’s, prospect theory is more flexible than it first appears.

Four Features of Prospect Theory

Kahneman & Tversky’s (1979) prospect theory has four features. First, reference points. The reference point
concept started life as zero or the status quo. Later, it was expanded to encompass goals or aspirations

Electronic copy available at: https://ssrn.com/abstract=4253679

(Lopes 1987, Lopes & Oden 1999, Heath, Larrick & Wu 1999). Unlike expected utility theory, prospect theory
says that the decision-maker might frame an otherwise positively valued outcome as a loss if it is below the
reference point, and vice versa. More important, reference points can be shaped or even determined by the
context. A decision-maker might frame his poor investment returns as a gain if they exceed the returns earned
by his friends. A decision-maker might perceive her annual bonus as a loss because it falls short of her
aspiration for the year or because it is less than her closest colleague’s bonus. An entrepreneur may disregard
projects in certain business sectors, not because they don’t promise to yield positive payoffs but because
those positive payoffs are not sufficient to permit him to reach or exceed his goals (which might, in turn, have
been shaped by a rival’s achievements).
Second, different preferences for risk in the domain of gains and domain of losses. The reference
point delineates a ‘domain of gains’ (points above the reference point) and a ‘domain of losses’ (points below
the reference point). This is illustrated in Figure 5, which shows the S-shaped value (utility) function inflecting
through the reference point. In the domain of gains, the decision-maker is risk averse, preferring to consolidate
accumulated or anticipated gains. In the domain of losses, driven by loss aversion, the decision-maker is risk
seeking and willing to take his or her chances to recover lost ground or improve the chance of securing an
outcome above the reference point. The higher the reference point relative to the average outcomes that the
available alternatives offer, the more risk the decision-maker must bear. If an investor’s reference point is a
20 percent return, for instance, then only relatively risky investments will permit that reference point to be
reached and surpassed. Loss aversion, depicted in Figure 5 by a steeper slope for the value function in the
domain of losses, reflects the fact that losses cut more deeply, more quickly. Loss aversion motivates the
decision-maker to avoid the domain of losses or get out of it if he or she happens to be in it.

Figure 5 The Reference Point, Domain of Gains, and Domain of Losses

Third, diminishing sensitivity. Changes in outcomes, 𝑥’s, a long way from the reference point do not have as
great an influence on the decision-maker’s valuation of an alternative as do changes in outcomes that are
closer to the reference point. Suppose that a company’s management team is weighing up two potential
acquisitions. Their reference point is an internal rate of return (IRR) of 20 percent. Each acquisition is subject

10

Electronic copy available at: https://ssrn.com/abstract=4253679

to risk and uncertainty, and each has a range of possible outcomes (IRRs). If, for one of the alternatives, one
of the outcomes that lies far from the reference point (e.g., an IRR of 120 percent) changes by 10 percent, this
is unlikely to affect the evaluation of that alternative by any significant degree. This a formal way of saying
that some outcomes are off the decision-maker’s radar and, as such, less attention is paid to them. Attention
is a scarce resource, and the decision-maker allocates it to outcomes (and changes to those outcomes) that
are close to the reference point.
Last, probability weighting. To compute the expected utility for an alternative, you sum the product
of the 𝑥’s and their associated 𝑝’s. Prospect theory incorporates an observation about how humans tend to
treat probabilities. They overweight unlikely outcomes and underweight more likely outcomes. In prospect
theory, therefore, the relative 𝑥’s (relative to the reference point) are not multiplied by their associated
probabilities but by probability weights, 𝜋’s. These weights add something to the lower range of probabilities
and subtract something from the higher range. Where expected utility requires an 𝑥 to be multiplied by its
associated probability of, say, 0.05, prospect theory might require the same 𝑥 to be multiplied by 0.07,
because people generally overweight unlikely outcomes. At the other end of the distribution, where outcomes
are more certain, expected utility might require an outcome to be multiplied by 0.90 whereas prospect theory
requires that same outcome to be multiplied by 0.86. Attention plays a role here too. Attention is drawn to the
far ends of the probability distribution, to zero and one. This gives us another S-shaped function. This time an
inverse S-shaped probability weighting function drawn in Figure 6.

Figure 6 The Inverse S-Shaped Probability Weighting Function

Innovation is always significant for those people closely connected to the most disrupted businesses, but the
resulting developments do not always spill over into other fields or into the public consciousness. Sometimes
though, entrepreneurs achieve a certain ‘transcendence’ either through their personality, business
achievements, or the products and services they have created or reorganised or rejuvenated. When this
happens, their ideas or achievements can become reference points for aspiring entrepreneurs or, more
disruptively, render existing reference points obsolete in fields of endeavour that are seemingly far-removed
from their own. When entrepreneurs move from field to field, the ripple effect of their actions can be felt even
more widely, possibly beyond business enterprise and into popular culture. Elon Musk is one such individual.

11

Electronic copy available at: https://ssrn.com/abstract=4253679

The question is whether our theoretical frameworks can contain the type of behaviour, often unexpected,
sometimes outrageous, that people like Musk exhibit.

Entrepreneurial Decisions & Decision Theory

As far as orthodox decision theory is concerned, there’s only one way to represent entrepreneurial decision-
making and that’s as an information gathering and utilisation phenomenon. The entrepreneur is better or faster
or both at search and judgement. Behavioural economics, on its own, doesn’t have much more to add. It is, in
its usual interpretation, a generalisation of orthodox decision theory that gives us some explanations as to
why people do not always maximise expected utility. However, while these interpretations are restrictive, the
way in which the decision problem is structured by both orthodox and behavioural theory gives us something
to grasp and something to build on. The problem space underlying even so distinct a theory as Kirzner’s
‘alertness and discovery’ can be represented as open and expanding in contrast to the closed and fillable
populated problem space of orthodox and behavioural decision theory. And, as we explained earlier, those
parts of Austrian economics that are focused on entrepreneurial judgement can be expressed in the language
of alternatives, 𝑥’s and 𝑝’s. All the pieces, however, do not currently fit together perfectly and it might be a
while before a coherent picture begins to emerge.
With X.com, SpaceX, and Tesla, Musk had resources available from previous business ventures. The
sale of Zip2 helped establish X.com and his share of the sale of X.com/PayPal helped establish SpaceX and
his control over Tesla. Cryptocurrency and Twitter were part of a scene into which Musk has entered at a
much more mature stage of his career. In each case, there was also something to work with. Online banking
had been evolving for two decades prior to the founding of X.com. The space industry was primarily state
controlled but there were price signals conveying important information (cost per pound of materiel). Tesla,
as one of its founders (Martin Eberhard) has stated, is a tech company focused on batteries, software, and
motors (Welch 2007). It was not out of place in Silicon Valley. In each case, the problem space was open and
in no considerable way populated with 𝑥’s and 𝑝’s. However, nor was there a completely blank space
confronting Musk in any of these businesses. Uncertainty but not total uncertainty. How was he able to
proceed, successfully thus far, into this uncertainty when others did not, could not or would not?
One thing that becomes clearer as Musk’s decisions are analysed in detail is that one needs (at least)
a combination of effectual and causal reasoning to explain them. Either one taken separately, even if it is
followed by the other, does not capture the essence of Musk’s choices. Once Musk had the capital from the
sale of Zip2, the chain of subsequent decisions can be partly explained by effectual reasoning. We could say
that his initial steps towards delineating the problem space in each case began with delineating a set of
alternative uses to which he could put his accumulated resources. We might be tempted to conclude,
therefore, that once he had made the commitment to each alternative allocation of existing resources, he
proceeded to determine which problems would be worth solving in each business context. However, this
distinct two-phase reasoning process does not explain Musk’s choices. With SpaceX, the cost of launching
materiel into space was prohibitively high and solving that cost problem was an imperative. The effectual
reasoning process that might explain the establishment of SpaceX cannot be separated from the causal
reasoning process that identified an outcome that would be worthwhile achieving. Even the much later
decision to establish Starlink, though more quintessentially causal in its underlying reasoning process, could
also be seen as the outcome of an effectual reasoning process focused on determining possible uses for the
growing amount of capital on hand (or available from investors) at SpaceX.

12

Electronic copy available at: https://ssrn.com/abstract=4253679

 When one attempts specific applications of this type of theory there is a feeling of something
essential slipping through one’s fingers. Behavioural economics, especially prospect theory, gives us a few
more things to hold onto, a few more things to help the application ‘stick’. The main task is to provide an
explanation of how an entrepreneur proceeds in the face of uncertainty. As anyone interested in navigation
will tell you, reference points are crucial. Lourtie & Rendas (1994) describe an underwater autonomous
navigation problem where detailed information is only available around sparsely distributed reference points.
At each reference point there is clearer information (e.g., precise coordinates concerning obstacles). In
between, however, there is less precise information. The problem is to use a combination of reference points
and ‘feel’ (sonar) to traverse the space in between reference points. The point is that this can be
accomplished. This leads us to suggest that underwater navigation by sparse reference points might form the
basis of a useful analogy.
The prospect theory reference point can be a goal or aspiration and, as such, is not a tangible piece
of reality. But it has some connection to reality. Unless the decision-maker is a hopeless dreamer, the
reference point provides something against which to assess possible outcomes. While prospect theory is not
dynamic in its original exposition, there is nothing stopping the reference point from changing and there is
nothing preventing the theory from incorporating multiple reference points (e.g., Folger 1984, Kahneman 1992,
Phillips & Pohl 2021, especially, Ch.10). As such, a single reference point can act as a tether as the decision-
maker wades out into the uncertainty and a series of reference points can act as a set of waypoints to guide
the decision-maker’s effectual/causal reasoning as he or she navigates through the uncertainty, in a manner
analogous to the underwater navigation problem. At the beginning, these reference points might be very
vague. A reference point to tether effectual reasoning might emerge from the way that another entrepreneur
has used (or has proposed to use, even hypothetically) the spare resources at another company. A reference
point to tether causal reasoning might take its shape from the outcomes that have been experienced (or
hypothesised) from solving a problem that is related or analogous to the one that the entrepreneur presently
seeks to solve. SpaceX would solve a problem. So would Starlink. So would…
Causal and effectual reasoning do not tell us anything about the attitudes towards risk and
uncertainty that might shape the decision-making processes they help us describe. Navigating uncertainty, of
course, is not for the fainthearted. Entrepreneurs might be described as people who are more willing to take
risk and embrace uncertainty. Whereas expected utility theory is predominantly cast in terms of degrees of
risk aversion, the prospect theory decision-maker is risk averse in the domain of gains and risk seeking in the
domain of losses. Gains and losses can be accumulated (realised) or expected. For example, the investor who
has lost $1,000 is in the domain of losses if his or her reference point was zero2 but so is the investor who has
only $10,000 now but who has set a goal (an aspirational reference point) of accumulating $1,000,000. The
investor who has made $1,000 when his or her reference point was zero is in the domain of gains but so is
the investor who has set an aspirational reference point rate of return of 10 percent and whose current
investment opportunities all have a minimum 10 percent rate of return.
If an entrepreneurial process involves a series of reference points and each reference point is
aspirational, the entrepreneur works out of a domain of losses at each step along the way, continuously
motivated to take risk to exit the domain of losses by reaching the next reference point. Elon Musk found
himself in the domain of losses at various times with each of his ventures. His newest idea, the Twitter

2It must be remembered that a loss of $1,000 might be considered a gain if the investor’s reference point was a loss of
$2,000.

13

Electronic copy available at: https://ssrn.com/abstract=4253679

takeover, if successful will certainly see him begin life as Twitter owner deep in the domain of losses.
Extricating himself and the company from that position and reaching the next reference point (e.g., a
profitability milestone goal) is motivating because of the discomfort that being in the loss domain brings. This
provides a theoretical foundation to the sense that we have of entrepreneurs as risk-takers without assuming
that such is some sort of unique or magical characteristic separating the entrepreneur from mere mortals.
What of the other two features of prospect theory? Diminishing sensitivity says that outcomes a long
way from the reference point do not play as much of a role in the ordering of alternatives as outcomes closer
to the reference point. The concept reinforces the centrality of the reference point as the decision-maker’s
focal point. Attention is drawn to the reference point. Diminishing sensitivity has not received a great deal of
attention in behavioural economics but like the other features of prospect theory it might one day be extended
to encompass relevant aspects of the decision-making behaviour. Regarding the entrepreneur, we might
simply agree with standard prospect theory and conclude that the entrepreneur focuses attention on each
reference point. Experiments might even reveal more laser-like attention among entrepreneurs. Or perhaps
the opposite. That is, the entrepreneur might be able to visualise a broader range of outcomes (𝑥’s) than
‘ordinary’ decision-makers. If so, the entrepreneur would find it possible (where others would not) to establish
multiple reference points across open areas of uncertainty.
Probability weighting is something we must approach with caution. If we are dealing with uncertainty,
there are no probabilities. Unless the entrepreneur has been able to fill in, subjectively, a set of probabilities
that sum to one, there is not much that can be drawn from the inverse S-shaped probability weighting function
that might be used to shed any light on entrepreneurial judgement. Even so, there is one interesting feature
of probability weighting that might be useful. Part of the reason why the weighting function is inverse S-shaped
is because the two extremes, zero and one, attract the decision-maker’s attention. In a sense, prospect theory
can be said to posit three reference points. The outcome reference point and two probability reference points
(𝑝 = 0, 𝑝 = 1). Even if the entrepreneur cannot determine a closed set of outcomes and a set of associated
probabilities that sum to one, the entrepreneur might be able to say, in a manner reminiscent of Shackle’s
(1961) potential surprise (non-distributional) measure of uncertainty, what is certainly impossible or certainly
possible (though not certain in the sense of 𝑝 = 1). In this way, some structure can be added to our
understanding of the way in which the entrepreneur perceives uncertainty. Again, it might be found that
entrepreneurs have a better sense of what is possible. Regardless, casting one’s imagination out into the
unknown is not an exercise in pure fantasy. There is something that ties the entrepreneur’s ideas about a
business opportunity to reality. It need not be a probability distribution, but some sense of the possible that
attracts and focuses the entrepreneur’s attention. Decisions made by people like Musk might be more clearly
understood by adding behavioural economics to the theoretical mix. And by challenging behavioural
economics to encompass the fundamental reshaping of the problem space that prominent entrepreneurs
bring about, we expand the reach of behavioural economics. Either that, or it breaks.

Concluding Remarks
Elon Musk’s online-media personality, part crafted by himself, part crafted by critics, part crafted by those
who idolise him, makes him stand apart, even in the very select crowd of entrepreneurs that dominate the
contemporary consciousness. Beyond this, there is a person making business decisions and we want to know
more about this decision-making process. If we use orthodox decision theory to examine these types of
decisions, we are led to conclude that there must be something superior about the entrepreneur’s search and

14

Electronic copy available at: https://ssrn.com/abstract=4253679

judgement processes because there is nothing especially unique about the problem spaces they face. Since
the problem space is closed and amenable to being filled or delineated by 𝑥’s and a set of associated 𝑝’s that
sum to one, it is only speed or superior intellect that separates the entrepreneur from ordinary people. Given
enough time and sufficient monetary and mental resources to gather the information, anybody can fill in a
problem space and then, even if they are not very good decision-makers inherently, they could apply the
expected utility theory to the set of alternatives and obtain the optimal ordering. If you have the full set of
alternatives, 𝑥’s and 𝑝’s this is rather easy.
Not surprisingly, both before and after the emergence of the economics of information, people looked
elsewhere for an explanation of entrepreneurial decision-making. The exceptions, of course, are the
neoclassical economists who are satisfied with an explanation based on imperfect information and search.
The Austrian School has traditionally offered the most prominent alternative to the neoclassical description
of the entrepreneur, but there is no single Austrian perspective. We can very roughly separate the Austrian
perspectives into (1) a Kirzner ‘discovery/opportunity/alertness’ perspective; and (2) a ‘reasoning and
judgement’ perspective. The reasoning and judgement perspective, reflected in such papers as Packard, Clark
& Klein (2017), is something that is more familiar and more understandable to orthodox decision theorists,
even if they find some of the ideas a little bit strange. However, the imperfect information and search
explanation is quite alluring, and many orthodox decision theorists are reluctant to embrace ‘stranger’ theories
when, to them, the idea that an entrepreneur is simply better and faster seems to encompass the essential
elements of the entrepreneur’s role in the economy.
It might be expected that behavioural economists would be more receptive to unorthodox ideas. The
problem is that behavioural economics is usually deployed within the same type of problem space as orthodox
decision theory. That is, closed and able to be fully populated by search. The main innovation is that
behavioural economics offers various explanations for divergences from optimal search and judgement. But
behavioural economics, including prospect theory, is far more flexible than its expected utility theory
precursor. And while the steps have been very gradual, the original version of prospect theory has been
extended in some interesting directions, demonstrating along the way how it can be used to better understand
decision-making in contexts that were not envisaged by Kahneman & Tversky. A good example is the way in
which the reference point has evolved from being interpreted as the status quo or $0.00 to a goal or
aspiration. Developments such as this then permeate through the theory. If the reference point is a goal, the
decision-maker approaches it from the loss domain. The discomfort of loss aversion motivates the decision-
maker towards the goal. Achieving a goal can be demotivating. The pain of loss aversion is gone and the risk
averse tendency to consolidate gains takes hold. There is a possibility, then, that behavioural economics might
be a pathway to connect the judgement theory of entrepreneurial decision-making with the more familiar
structures of decision theory. From there, more connections can be made to other theories, including other
theories developed by the Austrian School.
A series of decisions like Elon Musk’s demonstrate what we all know. Entrepreneurial decision-
making is hard to nail down. Entrepreneurial decisions are more than the result of faster and better information
gathering and judgement. More than an exercise in either causal or effectual reasoning. More than alertness
to opportunity. More than a greater tolerance for risk taking. More than swashbuckling adventurism with a
mixture of good fortune. There is something solid that underlies entrepreneurial decision-making. There is a
problem space, just not the closed and fillable one of decision theory, the one that underpins the neoclassical
treatment of the entrepreneur. And there is some sort of process that the entrepreneur uses to navigate into

15

Electronic copy available at: https://ssrn.com/abstract=4253679

and through the uncertainty. An Austrian explanation cast in the more familiar language of problem spaces,
of alternatives, 𝑥’s and 𝑝’s, making use of concepts such as causal and effectual reasoning and
complementing them with reference points, diminishing sensitivity, and probability weighting (as a start), may
be the best way to gradually fill in a picture of what the entrepreneur is doing when he or she steps out of the
closed problem space into the uncertainty beyond and how, by doing so, what was once uncertain becomes
knowable. Understanding how decisions by people like Elon Musk ripple through the decision-making
processes of others, changing problem spaces, changing reference points, changing what is perceived to be
possible, is another potential reward offered by this ongoing research program.

References
Andrejczak, M. 2010. Tesla Motors Revs Up $244 Million IPO. MarketWatch, June 28.
BBC News 2022. Elon Musk’s Tesla Sells Most of its Bitcoin Holdings. July 21.
Bielski, L. 2000. Online Banking Yet to Deliver. ABA Banking Journal, 92, September.
Boudette, N.E. 2022. Tesla Reports Record Yearly Profit but Warns That Supply Problems Persist. New York
Times, January 26.
Bursztynsky, J. 2022. Tesla Said It Held Nearly $2 Billion Worth of Bitcoin at the End of 2021. CNBC, February
7.
Business Insider 2021. US Digital Banking Users Will Surpass 200 Million in 2022. May 11.
Chow, D. 2022. To Cheaply Go: How Falling Launch Costs Fuelled a Thriving Economy in Orbit. NBC News,
April 9.
Coinbase 2022. Will Elon Make Twitter a Web3 Hub? April 27. https://www.coinbase.com/bytes/archive/will-
elon-make-twitter-a-web3-hub
Cronin, M.J. (Ed.) 1998. Banking and Finance on the Internet. John Wiley & Sons, New York, New York.
Davies, D. & Eltzroth, C. 2022. Crypto Boost May be Focus for Elon Musk as He Acquires Twitter. Bloomberg,
May 20.
Frank, R. 2022. Elon Musk Unloaded $22 Billion of Tesla Stock and Still Owns More Now Than a Year Ago.
CNBC, February 15.
Folger, R. 1984. Perceived Injustice, Referent Cognitions, and the Concept of Comparison Level.
Representative Research in Cognitive Psychology, 14, 88-108.
Foss, N. & Klein, P. 2012. Organising Entrepreneurial Judgement: A New Approach to the Firm. Cambridge
University Press, Cambridge, UK.
Heath, C., Larrick, R.P. & Wu, G. (1999). Goal as Reference Points. Cognitive Psychology, 38, 79-109.
Kahneman, D. & Tversky, A. (1979). Prospect Theory: An Analysis of Decision Under Risk. Econometrica, 47,
263-291.
Kahneman, D. 1992. Reference Points, Anchors, Norms, and Mixed Feelings. Organizational Behaviour and
Human Decision Processes, 51, 296-312.
Kirzner, I.M. 1973. Competition and Entrepreneurship. University of Chicago Press, Chicago, IL.
Kirzner, I.M. 1985. Discovery and the Capitalist Process. University of Chicago Press, Chicago, IL.
Kirzner, I.M. 1997. Entrepreneurial Discovery and the Competitive Market Process: An Austrian Approach.
Journal of Economic Literature, 35, 60-85.
Klein, P.G. & Bylund, P.L. 2014. The Place of Austrian Economics in Contemporary Entrepreneurship Research.
Review of Austrian Economics, 27, 259-279.
Lewis, H. 2000. E-banks Scramble for Your Business. bankrate.com.
Lopes, L.L. (1987). Between Hope and Fear: the Psychology of Risk. Advances in Experimental Social
Psychology, 20, 255-295.
Lopes, L.L. & Oden, G.C. (1999). The Role of Aspiration Level in Risky Choice: A Comparison of Cumulative
Prospect Theory and SP/A Theory. Journal of Mathematical Psychology, 43, 286-313.
Lourtie, I.M.G. & Rendas, M.J. 1994. Autonomous Navigation with Sparse Reference Points. Proceedings of
OCEANS’94, 3, 415-420, Brest, France, September.
Molla, R. 2021. When Elon Musk Tweets, Crypto Prices Move. Vox.Com, June 14.
Nath, R., Schrick, P. & Parzinger, M. 2001. Banker’s Perspectives on Internet Banking. e-Service Journal, 1, 21-
36.
Packard, M.D., Clark, B.B. & Klein, P.G. 2017. Uncertainty Types and Transitions in the Entrepreneurial Process.
University of Omaha Marketing and Management Faculty Publications, 17.
Phillips, P.J. & Pohl, G. 2021. Behavioural Economics and Terrorism: Law Enforcement and Patterns of
Behaviour. Routledge, Abingdon, UK.
Root, A. 2021. How SpaceX and Tesla Could Make Elon Musk a Trillionaire. Barron’s, October 20.

16

Electronic copy available at: https://ssrn.com/abstract=4253679

Servon, L.J. & Kaestner, R. 2008. Consumer Financial Literacy and the Impact of Online Banking on the
Financial Behavior of Lower-Income Bank Customers. Journal of Consumer Affairs, 42, 271-305.
Shackle, G. L. S. 1961; 1969. Decision, Order and Time in Human Affairs, Cambridge University Press,
Cambridge.
Sheetz, M. 2021. Elon Musk’s SpaceX Raised $850 million, Jumping Valuation to about $74 Billion. CNBC,
February 16.
Sheetz, M. 2022a. SpaceX’s Starlink Satellite Internet Surpasses 400,000 Subscribers Globally. CNBC, May
25.
Sheetz, M. 2022b. Elon Musk Says an IPO of SpaceX’s Starlink Satellite Internet Business is Still 3 or 4 Years
Away. CNBC, June 7.
Shevlin, R. 2020. The Online Bank Insurgency of 2020. Forbes, July 20.
Strohm, M. 2021. Digital Banking Survey: 76% Of Americans Bank Via Mobile App—Here Are The Most And
Least Valuable Features. Forbes Advisor, February 24.
Tan, G. & Roof, K. 2022. SpaceX Employees are Quietly Selling Shares at a $125 Billion Valuation. Fortune,
May 17.
Tong, K. 2022. How Relevant is Twitter to Most People? ABC News (Australia), April 27.
Thomas, O. 2010. Tesla’s Elon Musk: I Ran Out of Cash. VentureBeat.com, May 27.
Wasson, S. 2020. The Big Goodbye: Chinatown and the Last Years of Hollywood. Faber & Faber, London, UK.
Welch, D. 2007. Tesla: A Carmaker with Silicon Valley Spark. Bloomberg Businessweek, July 30.
Winkler, R. 2018. Elon Musk’s SpaceX Is Raising $500 Million in Funding. Wall Street Journal, December 18.

17

Electronic copy available at: https://ssrn.com/abstract=4253679