Economic models describe a world that does not exist: everyone knows everything, trading is free, prices move at once. So does any of it work outside the classroom? Make your guess first.
Partly right. Models often get the direction of an effect right, such as higher interest rates cooling spending, while their exact numbers rarely hold. Keep reading to see where they break.
Fair: the assumptions are not literally true. But a map that showed every tree would be as large as the forest. Keep reading to see what the simplification buys.
The economist's answer, and the right one. It depends on which assumption matters for the question you ask. Below you can break them one at a time.
The world inside the model.
A market in equilibrium quietly assumes a list of things. Switch three of them off and watch a simple market move away from its ideal.
Break an assumption, watch the market
Everyone knows the quality and price of everything they trade.
A used-car buyer cannot see the engine; a lender cannot see how careful a borrower is. When buyers cannot tell good from bad, they pay for the average, and good sellers leave. Portfolio Choice assumes every investor knows each asset's expected return and risk.
Buying and selling are free and instant.
Brokers charge fees, houses take months to sell and changing jobs costs time and money. Each cost drives a wedge between what buyers pay and what sellers keep. Portfolio Choice lets you buy and sell any mix of assets for free.
Many buyers and sellers, none large enough to move the price.
In many industries a few large firms set their own prices. A seller with power sells less, at a higher price, than a competitive market would. The New Keynesian model gives firms some power over their prices: it is why they choose when to change them.
Demand (true value)
Supply (cost)
What buyers will pay without information
Supply plus the cost of trading
Single seller's extra revenue
Gains from trade lost
Units traded
–
Buyers pay
–
Sellers keep
–
Gains kept
–
Made-up units. Demand is what buyers would pay for each extra unit; supply is what each extra unit costs to provide. The shaded area is the gain from trade the market loses compared with its ideal.
Model: one-good market with straight-line demand and supply (an illustration, not data).
And five more the models lean on
Tap one to see what real life does instead.
Menus, wages and contracts change only now and then, so a surprise can leave workers idle or shelves empty for a while.
The New Keynesian model drops this one: prices are sticky, which is why a central bank can move output.
People use rules of thumb, put off decisions and follow the crowd.
New Keynesian households and firms form their expectations rationally.
In the data many things move at once: in 2022 interest rates, energy prices and supply chains all changed together.
The Solow Scenario Lab changes one parameter at a time, on purpose.
Households differ in income, wealth and age; a rate rise helps savers and hurts borrowers.
The Solow and New Keynesian models both use one representative household.
Saving and growth drift over decades and jump in crises.
Solow calibrates each country with its averages over the period you choose.
If the assumptions break, what is the point?
Four reasons the theory earns its place, even though its world does not exist.
It isolates one force.
Holding all else equal is how you see what a single cause does, the way a laboratory holds the temperature fixed to study one reaction.
The direction usually survives.
Higher interest rates cool spending, more investment builds capital, more stocks lower risk. These hold even when the exact numbers do not.
The ideal is a yardstick.
Once you know where a perfect market would settle, the gap tells you what frictions cost and what a policy could fix.
It tells you when not to trust it.
A model is most dangerous when you forget what it assumed. Knowing the list is how you spot the case where it breaks.
Ceteris Lab
Economists say ceteris paribus, Latin for "all else equal", when they hold every other force still to study one. Real life never holds still. This lab starts from the theory and then lets you test what happens when all else is not equal.
That is how this website was built.
I wanted to connect theoretical economic and financial concepts to real-life scenarios. In class, a model ends when the equations are solved. Outside, that is where the questions start: does this happen in my country, with my savings, with this central bank?
So every model here comes in two halves. The theory is worked out exactly, from the same equations a university course uses. Beside it sits a real-life check: real data, a real decision or a real event, where you can see how far the theory goes and where it stops.
Patricio Elizondo
Theory meets real life.
Each model on the site, with its theory, a piece of the real world to test it on, and a question to keep thinking about.
Solow growth model
Why are some countries richer than others?
The theory
Countries that invest more of their income, and whose population grows more slowly, settle at a higher income per worker: their steady state.
Real life
Pick a country. The model reads its real investment, population and technology data and says whether it sits above or below its own steady state.
Investment share, s
–
Population growth a year, n
–
Technology growth a year, g
–
Versus its steady state
–
Model: Solow, calibrated on the country's full period. Data: World Bank, Penn World Table 11.0.
Did raising interest rates bring inflation down in Canada?
The theory
In the model, a surprise rise in the policy rate lowers inflation and output at once, and both drift back as the surprise fades.
Real life
In Canada, inflation rose for a year while the policy rate stayed at its floor. The Bank of Canada then raised it quickly, and inflation came down over the following two years.
What happened: Canada, quarterly
Inflation, year over year (%)
Policy rate (% a year)
2% inflation target
Data: Statistics Canada, Bank of Canada.
What the model says after a surprise rate rise
The three years after the rise, quarter by quarter
Inflation–basis points per yearOutput gap–percent of natural output
Model: closed-economy New Keynesian model, quarterly.