Maya J. Lewis - "Bet You $100 Elon Will Headline the Superbowl"

An American’s dopamine is worth its weight in gold – companies see most profit when they offer consumers even just the possibility of earning some kind of secret, though highly admired, reward, drawing attention and teasing their innate, alluring curiosity about “What if?” Consider Labubu’s blind box model. Didn’t get the plushie you wanted on your first purchase? Fear not, there’s always another chance! 

“Prediction markets intentionally complicate gambling’s common definition, blurring the line between betting and trading by transforming curiosity about ‘what if?’ into a price.”

In just seven years, increased gambling culture in the U.S. has exposed sports betting beyond Nevada, and into a virtual economy that manages $100 billion wagered annually on online sportsbooks. Since 2018, when the Supreme Court outlawed the Professional and Amateur Sports Protection Act, a 1990s law that prohibited sports gambling, 1 in 10 Americans report having placed at least one bet within the last year. Today, 35 states, alongside Puerto Rico and Washington, D.C., have legalized sports betting in some form, growing the market from a previously niche, taboo activity to a mainstream economic powerhouse. Yet, there’s a new player in town: prediction markets. These markets allow trading on outcomes as varied as sports results, Taylor Swift’s streaming numbers, or the length of Donald Trump’s inauguration speech and the exact words he’s expected to share. But how exactly do these new markets differ from the gamble of sports betting? 

Prediction markets intentionally complicate gambling's common definition. Platforms such as Kalshi and Polymarket allow users to trade against counterparties (other traders within the platform), rather than “the house,” similarly to stocks or futures. Furthermore, prices reflect the market’s assessment of probabilities, using a 0-100 scale. For example, 75 cents implies traders believe there is a 75% chance of that outcome manifesting, and a 25% chance that it does not. If someone bought 100 shares on a prediction, their profits would be calculated as the spread between $1 and the price at which the shares were purchased – the spread is then multiplied by the number of shares purchased at that time. This simulates a “trading” experience rather than traditional betting. 

While sports betting is regulated on a state-by-state basis, prediction markets exist at the federal level, overseen by a government agency known as the Commodity Futures Trading Commission (CFTC). It’s a sticky situation: legal barriers prohibiting the interstate transmission of bets and wagers mean that semantics have proven to be a crutch for the survival of prediction markets. In an effort to curb regulatory scrutiny, platforms argue that trades are merely “event contracts” and, to abide by federal law, these contracts cannot rely upon an outcome related to terrorism, assassination, war, or other federally illegal activity. Such platforms face cease-and-desist letters; nevertheless, engagement continues to skyrocket. Kalshi alone handled $208 million in March Madness bets this year.

Profitability in the country has surged: $11.04 billion in 2023, $13.71 billion in 2024, and a recent industry estimate projects the prediction market sector will grow to nearly $95.5 billion by 2035, reflecting annual growth rates approaching 47%. Since the Supreme Court’s decision, this demographic has shifted from illegal bookies, which once claimed a $150 billion annual market. Now, much of recent activity is federally regulated – 70% of online bets placed today are legal compared to just 44% in 2019, which highlights a dramatic transition toward a transparent and monitored industry. 

From an economic perspective, these platforms illustrate the inner workings of efficient markets. Traders incorporate public information, private insights, and probabilistic reasoning to demonstrate active sentiment on infinitely many events. Federally regulated markets enable transparent pricing, taxation, and oversight, contrasting with previously dominant illegal transactions. This lack of oversight deprived states of $700 million in annual tax revenue. 

Because prediction markets’ earnings are not collected on the basis of betting returns and their operation under CFTC oversight associates them closer to futures than gambling, the IRS does not treat these profits the way it would winnings from a sportsbook – they’re fundamentally different from sports betting, and are neither gambling winnings nor capital gains. Instead, it is considered “other income,” similar to money received from odd jobs or freelance work, and there is often a lower tax burden. Sports betting requires reporting winnings using Form W-2G, whereas prediction earnings experience no automatic withholding, no flat tax rate, and no gambling classification – another incentive to take advantage of prediction markets. 

The prediction markets, whether one chooses to directly associate them with gambling or not, have proven themselves to be functioning financial markets that aggregate dispersed knowledge and shape incentives, also revealing how individual decision-making under uncertainty interacts with broader economic and government structures. As regulation and technology continue to see outstanding intersections today and, likely, in the future, we should embrace more of the opportunities and challenges of incentive-driven decision-making between government and citizens in the digital age.

Audrey Mallier - Spotify Wrapped: Price-Discrimination in Disguise?

Like clockwork, every December millions of people worldwide are equally excited as last year to flood the internet with neon screenshots from their Spotify app. Top genres with hyperspecific categories catch your attention, such as “Underground Hip Hop,” “Vapor Soul,” and “Dream SMP,” among many others. Spotify Wrapped feels personal, fun, and understandable, serving as a way to connect with friends and show them a glimpse of your inner life. Yet, there’s something deeply calculated behind the confetti and bolded statistics. Wrapped isn’t simply a pop-culture sensation; it’s a trending example of modern data-driven price discrimination in a digital market. While Spotify Wrapped may just seem like a celebration of your commitment to music, it’s also evidence of how well Spotify understands your willingness to pay and profits from it. 

Leaving Spotify means losing your identity. That’s why Wrapped works. 

Price discrimination can be described in very simple terms as the ability of firms to charge different consumers different prices based on how much each is willing to pay. There are three degrees of price discrimination since in many markets, this requires information about consumers that firms have very little knowledge of: preferences, income, loyalty (Pearson, Microeconomics 3rd Edition Textbook, C. 12.) But one specific platform defies this. 

Spotify tracks every second of every song you listen to, every song and album skipped, your artist and genre preferences, and more, which allows it to make data-driven assumptions about when you concentrate, when you sleep, when you host parties. Not only does this powerful behavioral data translate to perfectly curated recommendations and personalized playlists, but also the ability to segment listeners based on their preferences. 

Spotify loves these revealed preferences. For example, if Wrapped shows that you have listened to 40,000 minutes of Lady Gaga, Spotify can say with near perfect confidence that your demand for Lady Gaga’s music is inelastic, meaning that you will keep listening to her music even if ad frequency increases or Spotify Premium prices rise. Additionally, you are a premium target for concert ads, merch links, or exclusive artist information. Suddenly your tastes become economically legible. Heavy users get prompted to buy Spotify premium, and algorithm dependent users, people who frequent the personalized playlists or “daily mix,” are targets for playlist experiences that keep them captivated. 

Instead of observable differences between consumers, Spotify differentiates by behavioral differences. Casual background music listeners suggest that those users have a low willingness to pay and elastic demand. Spotify knows that these users don’t value the service enough to pay for premium, so they monetize them through ads rather than premium incentives. People who only listen to workout playlists are daily users and have a higher willingness to pay and are more likely to pay for premium. Wrapped uses Richard Thaler’s nudge theory to make consumers internalize these identities. Using phrases such as “top 0.1% listener of Taylor Swift” makes listening to music more personal and switching platforms “costly.” If Spotify defines you as an “Indie Core listener,” you are more likely to act like one. This is a psychological cost that reduces competition from other listening platforms. 

Spotify users feel a little more seen when December rolls around. 

Beyond a feeling of satisfaction, your Wrapped becomes a signal of status. Screenshots are shared to group chats, posted on social media, and discussed among friends for validation. This creates a positive network externality for Spotify, increasing the app’s value when each person shares their Wrapped. 

Seeing the temporary “Wrapped” tab at the top of your Spotify app with fun images is an annual reminder that you’ve given Spotify the information it needs to design a revenue model based on your preferences. Wrapped works because it doesn’t just monetize your data; it also makes you feel good. 

Leaving Spotify means losing your identity. That’s why Wrapped works. 

With the same excitement as the year before, Spotify Wrapped continues to succeed because it transforms your preferences into a product. It’s a personalization that feels good, but functions as a dataset and strategy. Its economics There is no secret Spotify hides from its consumers, but its disguise as celebration is a party we all partake in nonetheless. 

Luke Shepard - Gambling for good?

SOPA Images/LightRocket via Getty Images

Prediction platforms have exploded in popularity in the past few years.

Polymarket, one of these platforms, saw over $3 billion in investment on the 2024 election alone. Public opinion surrounding these services has been rather controversial, with some lauding the potential benefits while many criticize these same platforms as just another method of losing money to a gambling service. As the benefits of investing (or gambling) on these platforms become the primary focus, prediction markets may serve as valuable forecasting tools for the general public. The search for private profit may inadvertently create a public good in a prediction market.

Prediction markets frequently provide predictions that are more accurate than those of pollsters or other professional forecasters. Prediction models, particularly in politics, have long suffered from a lack of profit incentive to motivate accurate results. Funding problems often limit the ability of agencies to survey widespread groups and then share accurate results. The most well-funded programs are typically private services which sell products to specific groups, preventing widespread accessibility for the public. Even publicly available polls are flawed prediction models due to time delays in interviewing survey candidates, general reluctance to respond to pollsters, amongst many other issues.

Prediction markets counteract these shortcomings by combining the public and private interests. Investors use these prediction markets in order to try to maximize their individual profit. Using whatever information at their disposal, their opinions of the market serve to affect the prices of the market. Prices match up perfectly with potential outcomes, where a 60% chance of an event occurring reflects a price of $0.60. Profit incentives also can help avoid potential biases associated with poll responses (Forbes).

The recent mayoral race in New York City exemplified the true capabilities of prediction markets. Polls consistently wavered with regard to which candidate proved most popular for NYC voters. As with all political polls, they took days to account for an accurate and general sample size and to try and avoid as much survey bias as possible. Prediction markets like Kalshi acted swiftly to predict a Mamdani win much more assertively since his selection as the candidate for the democratic party (Forbes). The effect was clear: prediction markets responded swiftly and accurately to the general public.

With increased accuracy, these prediction markets are then able to act as public goods. The prices that these investments create reflect to the public the chance that certain events could occur. Unlike individual predictions, these markets then become collections of information where individuals can access relevant predictions of the future direction of the economy. These markets cannot exclude individuals from viewing the prices/predictions of the consumer base as a whole as they intend to attract more investments. Thus, regardless of whether one seeks to invest on these platforms, they have access to prediction models that have frequently proved successful.

Prediction models provide a unique innovation in creating a public good out of the private market. For those who seek to make profit, they are able to hedge their predictions against market

predictions. Yet, for the public as a whole, they are given access to the agglomeration of these predictions from a variety of individuals. The benefits of these predictions can extend to a variety of aspects of private life. Prediction markets often focus on the odds of specific policy implementation, inflation rates/other economic indicators, and political outcomes. Given how future knowledge of these outcomes can help the public prepare in their personal and economic lives, prediction markets can serve as the most accessible, accurate metric for helping individual preparation for the future.

However, this may be too cheery of a description of these sites. Almost indubitably, many aspects of them serve as public goods. As related to a more formal economic definition, they are non-excludable goods in that the information they estimate regarding the future is open to the public. These goods are non-rivalrous as there is not a set amount of goods available, given that the market expands with additional investment.

Although there may be a valid interpretation of predictions markets as a public good for informing the public on potential decisions, these platforms are very dangerous for many users. As reported in “Gambling, Prediction Markets Create New Credit Risks, BofA Warns,” consumers are taking on large debts and defaulting on loans as a result of prediction markets. While it is valuable to acknowledge how prediction markets can act as public goods in offering accurate interpretations of potential future outcomes, these potential benefits come with consequences frequently associated with habits related to risk with money.

Edward Ward - TikTokification of Politics

Prompted | OpenAI

The average attention span is only 7 seconds long

This line has become inescapable in contemporary life, and almost invariably, TikTok, Instagram Reels, and YouTube Shorts are mentioned in the next breath. Supposedly, thedopamine hits from endless short-form content rewire the brain’s reward system so that no one is able to concentrate for an extended period of time. Whether or not this is true, it is unfortunately an apt metaphor for a particular aspect of American fiscal policy: the national debt.

Between 1975 and 2005, the American debt-to-GDP ratio was relatively low, averaging 36% during these years. In the twenty years since, the ratio has ballooned to the point where it is over 120% of GDP. The amount of money spent on maintaining said debt through interest payments now exceeds that spent on defense, and is only outpaced by Social Security and Medicare/Medicaid. To put it mildly, the nation has spent too much time scrolling.

Unfortunately, politicians such as Zohran Mamdani and President Trump seem to have no interest in addressing this frightening reality. Indeed, the Congressional Budget Office predicts that the Republican-backed Big Beautiful Bill will increase federal debt by nearly three trillion dollars over the next nine years. What happened to the party of fiscal responsibility?

The truth is that tax cuts and increased government spending are fiscal policies with particular appeal to Republicans and Democrats, respectively. From the perspective of our leaders, who need to worry about being reelected, it makes perfect sense to call for these measures. Their political futures, and therefore a non-zero amount of their private lives, are tied to their support for measures that are in turn supported by their voters. Why rock the boat with party elites and constituents at home by trying to ensure Social Security’s long-term solvency- a problem the country will face in the next ten years- when there’s an election in 2026? In other words, why not watch another TikTok?

I fear that this is a particularly sinister incidence of the principal-agent problem. In a republic like the United States, voters (the principal) elect representatives (agents) to govern in their stead and on their behalf. Theoretically, we elect the candidate who we believe will govern most in our interest. Elections, therefore, should act as an incentive for elected officials to govern in the interests of the public, rather than serve their personal, political needs.

Regrettably, this has not been and continues not to be the case. Whether this is the fault of special interests making representatives the servants of two masters, asymmetric information between voters and politicians, or some other reason is of little importance. Spending and taxation must be brought into alignment, and pro-growth policies have to be championed by our elected officials. The public at large must also do its part and think critically about what, more than who, they are voting for. Everyone has to put down their phones, so to speak.

This call for restraint with the national credit card is not meant to throw the most vulnerable among us under the bus. The debt taken on today will have to be paid for in some way, whether directly through taxes or indirectly through inflation, higher interest rates, or less spending on government programs. Some of this is already being seen. Inflation, which disproportionately hurts lower earners and recipients of fixed incomes, is still above the Federal Reserve’s target of 2.0% five and a half years since the start of the pandemic. The ten-year Treasury yields and thirty-year mortgage averages remain at some of their highest rates in the last twenty years, making a key milestone of financial security even more out-of-reach for young people across the country. Governments have a responsibility to look out for their most vulnerable and most productive citizens alike, and the federal government is not taking enough care as it relates to our long-term interests.

President Trump was reelected in no small part thanks to the economic concerns of Americans, and affordability continues to be a top concern for American voters. This is an opportunity for politicians, as they can reframe changes in policy regarding spending and taxation as steps towards a sound fiscal future. Perhaps more importantly are the gains in productivity that AI seems to promise us, which could allow us to partially grow our way out of the crisis by shrinking the debt-to-GDP ratio. None of this will be easy, and must be done carefully so that we do not place more pressure on a country struggling with the cost of living, but it is possible. We have to stop chasing political and economic dopamine, and focus on the longer, harder task of fiscal discipline.