One of my kids has recently discovered this gallery of personal finance games for kids, all of which deserve a tacit endorsement because while they do present players with realistic-sounding choices with a finance flavor, all of them fall down pretty hard in the "being a good game" department. One taxonomy of games, which applies to both video games and board games (and to a lesser extent sports) works like this:

1. A game can be a fairly simple set of rules which, in combination, can lead to surprising and unpredictable outcomes. It's not very hard to write down the rules of poker, chess, or go, but each time a computer has reached human-level play with these games, it's been a major milestone in the history of artificial intelligence.

2. Some combination of the above, coupled with a simulation. Magic: The Gathering is really a competitive programming game where each character gets a randomized set of functions, some of which can call each other and some of which can give them access to other functions. It's also a game where you pretend to be wizards fighting. In practice, advanced players seem pretty uninterested in lore, but lore, card art, etc. attract a wider audience, enabling Magic to both get players hooked when they're young and ensure a steady supply of opponents who don't run things strictly optimally. And this applies to many such games; the initial attraction is the simulation aspect, but people who put in the hours and crank up their skills are treating the simulation as one of many equivalent interfaces for the same abstraction.¹

3. Some games present themselves as a simulation that provides context around a set of simple rules, but actually serve a more didactic purpose. The game isn't saying "Imagine what would happen if you were in this situation," but saying "I will tell you precisely what will happen in this situation." So the crypto trading game in the gallery above will randomly zero coins, or have exchanges fail; Spent, a game about living paycheck-to-paycheck, is a game whose mechanics are set up so that it's impossible to win. (The closest you can get to a victory condition is that you make it to the end of the month without running out of money, and then get informed that next month's rent is due.) Payback, a game about paying for college, does give you a win condition, but it mostly involves making prudent middle-of-the-road choices—go to an affordable mid-tier school, study but don't study too hard, choose a sensible job, etc.

There's a case for teaching these lessons. The average prospective college student should think about how to balance their interests against financial reality, and they should be aware that if they're living paycheck-to-paycheck, it means that any unexpected costs come with unpleasant tradeoffs. But it's also true that this preachy approach makes the games worse simulations of reality, and frankly no fun at all. Spent is not interested in optimal financial planning for the hanging-on-by-their-fingertips class. After all, if it's truly the case that poverty is expensive, a necessary consequence of this is that people don't just quietly emerge from poverty, but rocket out straight into the middle class. If, for example, you were a few hundred dollars short for a medical bill, which caused you to miss your car payment, which got you evicted, which meant you were sleeping rough, which meant you got sick again, in an ever-downward spiral—then the first one-month emergency fund means you've suddenly escaped this dynamic, and have higher job security, better health, less stress, and more free time, all of which can be applied to many things including the project of getting more money. (A more succinct way of critiquing the "high cost of poverty" argument—which is true, but also not the main contributor to poverty—is that if it were the main cause, the most effective anti-poverty program would be the lottery. This is not so.)

Games can be fun when they're interacting with some system that plays by coherent rules and that also interacts with itself. This gives rise to interesting second-order consequences, which make the player's decisions more fun—it's the difference between a one-person play where the protagonist interacts with props and an ensemble cast where the protagonists are interacting with one another. Interestingly enough, one of the games in the list does offer this, in a way: Build Your Stax is a personal finance game that revolves around allocating money to different asset classes, like bank accounts, CDs, government bonds, etc.—but also to a selection of individual stocks. These stocks have fictional names, but their performance is based on actual historical performance for several high-profile companies. So in this case, instead of interacting with a simulated economy, players are interacting with the actual economy, and given the amounts involved these interactions are realistic.²

Good games are the ones that are breakable, and in interesting ways. There are a few ways to achieve this:

1. Multiplayer tends to solve for the "interesting ways" part quite well, because other players are implicitly trying to reverse-engineer your strategy and tactics in order to counter them.

2. In a single-player context, some combination of smart AI, numerous simple AIs whose emergent behavior looks smart, and nonlinear barriers to scaling keep things fun. (Paradox's grand strategy games often do this. If conquering twice as much means you square the number of ensuing problems, there's a lot of fun to be had at being at the edge of optimally-breakneck expansion—and this also means that any random surprises can lead to very interesting outcomes, and that advanced players are frequently making calculated gambles.)

One reason games don't focus on this is that it's not really their purpose. An economic simulation is fun when narrowly scoped, but a lot of profitable activities are profitable because they're either high-risk or kind of tedious. And if they're neither, they're probably outside the scope of whatever the game can provide: if you made Sim Economy in early 2020, how would you know to make generative AI possible? One of the better in-game market simulations in a single-player game is Railroad Tycoon II, which features a 19th century-style stock market where AIs will try to squeeze shorts or try to short high-flying stocks until their owners get margin-called. And the value of those stocks is roughly tied to the underlying companies—companies that can also do buybacks or issue shares on the secondary market. Unfortunately, it's just robust enough that with a little practice the financial side of the game comes down to beating up on AI players, which gets tedious after a while.

In a way, it's a validation of the importance of the market that it's hard to make a simulation that feels true to life. There are detailed market simulations that produce realistic price movements, but these aren't monetized in games—that's just a description of what systematic hedge funds do. You can make a game that teaches a specific lesson about money, but it's hard to track down a general-interest game that gets all of this right. The closest you can find is probably a very addictive mobile app I've logged quite a few hours of my time in; it's called Interactive Brokers.³

Read More in The Diff

This piece is at the intersection of games and money, one that’s been covered by The Diff from several angles. We’ve looked at: 

• How Factorio changes the way people think, and why Shopify lets employees expense it.

• How games benefit from the economics of time ($).

• The difficulty of RobinHood’s casual game/financial institution mix ($).

The surprising parallel between early-career interest in arbitrage and a youthful addiction to competitive gaming, and how both contribute to (or at least correlate with) later success.

Join the discussion!

1  The Civilization series lampshades this a bit, the basic game mechanics of cities and units remain fixed for the duration of play, while the amount of time represented by a single turn goes from decades at the start to one turn per year at the end. This actually ends up simulating both the increase in transportation speed from prehistory to the present and the rapid acceleration of economic growth throughout the last few centuries. Take it too literally, and it gets silly again—aircraft carriers taking half a decade to cross the Pacific, bombers returning from their mission after a two-year flight, etc. But it's directionally right, and making the details accurate means playing a more detail-oriented game—if you want reasonable timelines for modern military conflicts, you probably want the Hearts of Iron series rather than Civilization, but HOI also simulates a single conflict, and even then the pre-conflict parts are quite dull because any mechanic that would make them interesting would be all too distracting during wartime.

2  One consequence of choosing real-world companies is that, for gameplay reasons, they'd want to choose companies that had roughly similar performance to the overall market. They don't want the lesson to be that stocks can go to zero or that if you pick the right one you do really well, so that makes sense in light of the lesson they're teaching. But it also means that a mean-reversion strategy works; these stocks are more volatile than the index, but end up in the same place overall, so they went through cycles of out- and under-performance. Still, they look pretty realistic.

3  Every game has a metagame, and in the interest of full disclosure I'm playing both—that's a referral link.