Models
In development Ramsey–Cass–Koopmans growth · Search and unemployment · Intertemporal foundations · Money and banking · Forecasting
Explore how a different assumption changes an economy’s journey.
Change one assumption and keep everything else the same. Both paths start from the country's capital in the last year you selected, so any difference between them comes only from your change.
The baseline uses the averages of the years you chose. The years shown afterwards are what the model implies after the change, not a forecast.
Shares and annual growth rates are shown as percentages.
Load a baseline to start exploring.
Loading baseline…
A small change. A different future.
Choose a country, adjust one assumption, and compare the journey with the unchanged baseline.
Long-run levels are measured per unit of effective labor: labor adjusted for technology, so they leave out the steady rise that technology brings to living standards.
Output per unit of effective labor
Where investment meets break-even investment, capital per effective labor stops changing.
This experiment uses the continuous-time Solow model, with technology measured from each country's data so that the model matches every observed year.
Output per effective worker: y = kα
How capital changes: dk/dt = s kα − (n + g + δ) k
Steady state: k* = [s ÷ (n + g + δ)]1 ÷ (1 − α)
How much should an economy invest for its highest long-run consumption?
Investing more builds more capital and more output in the long run, but it leaves less to consume today. The Golden Rule is the investment share that gives the highest consumption once the economy has reached its steady state.
Pick a share to see the trade-off: how consumption changes right away, during the adjustment, and in the long run. In this model the investment share plays the role of saving.
The investment share stands in for saving in this Solow model.
Load a country to begin.
Loading country data…
A question about today and tomorrow
Choose a share to see how consumption changes immediately, during adjustment, and in the long run.
All levels are output available for consumption per unit of effective labor, not measured household spending.
Output per effective worker: y = kα
Consumption: c(t) = (1 − s) y(t)
How capital changes: dk/dt = s kα − (n + g + δ) k
How quickly can an economy close its steady state gap?
An economy away from its steady state moves toward it. Close to the steady state, each year closes about the same share of whatever gap is left, so the last part of the journey is the slowest.
Choose how much of the starting gap should close. The model computes exactly how many years that takes for the selected country, and also shows a common shortcut that works well when the gap is small.
50% closed · 50% remaining
Loading countries…
A question about adjustment
Select a country and milestone to see its starting gap and the model's adjustment time.
Capital per effective worker changes by dk/dt = s kα − (n + g + δ) k
Speed of convergence near the steady state: λ = (1 − α)(n + g + δ)
Turn a historical period into one Solow phase diagram.
The phase diagram is the heart of the Solow model. The curve shows how much the economy invests at each level of capital per effective worker; the straight line shows how much investment is needed just to keep that capital from shrinking.
Where the curve is above the line, capital grows; where it is below, capital shrinks. The point where they cross is the steady state. The years you choose set the investment share and the growth rates used to draw both.
Loading countries…
Two curves, one destination
Build a country model to compare actual investment with break-even investment and see where the economy sits relative to its model steady state.
The horizontal axis shows possible capital levels, not years. The investment curve flattens because each extra unit of capital adds less output when population and technology are held fixed.
These levels divide capital and output by population adjusted for productivity. At the steady state they stop changing, while output per person can still grow with technology.
Plot the real data and place it beside benchmarks.
These are the historical numbers the Solow model is built from: output, investment, population, capital, technology, and the investment share.
Put a country beside the average of a fixed group of countries, the lowest and highest countries of that group, or a random country. The differences you see describe what happened; on their own they do not explain why.
Every number the model uses comes from real data. Pick a country and a series to see its history.
Every number the model uses comes from real data. Pick a country and a series to see its history.
See where a country stands and where the Solow model says it is heading.
The Solow model says every economy is pulled toward a steady state: the level of capital per effective worker at which new investment exactly replaces what is used up by depreciation, population growth, and technology growth.
This summary measures where the chosen country stood in the last year of the period, where its steady state lies, and how long the model says the trip takes. The steady state is a benchmark built from the period's averages, not a forecast.
Apply a country and historical period above, then select Generate report to explore its position, long-run levels, and convergence.
Current position vs. steady state
Capital per unit of effective labor
All levels are measured per unit of effective labor.
Follow one shock through the four markets of the New Keynesian model, step by step.
The New Keynesian model in its simplest form. Prices were set in the past and cannot move this period, and the central bank sets the interest rate. So output is whatever is demanded at that rate. Firms hire the workers that output needs, the wage is whatever attracts those workers, and the central bank supplies the money that keeps its rate.
Pick a shock to see which curves move and how the effect passes from one market to the next. No real data is used: the curves come from one illustrative set of equations, listed under "Assumptions and the math".
Loading the model…
Prices are stuck. Demand decides.
Choose a shock, then follow it from the goods market to the money market.
Before After Before the exchange rate moves x Shocked variable x Changed by an earlier step
| Variable | Before | Change | After |
|---|
See how output, inflation, and interest rates react to a surprise when prices adjust slowly.
This is a small model of the short run with three relationships: how spending responds to interest rates, how inflation responds to the state of the economy, and a rule the central bank follows when it sets its rate.
One surprise hits in the first quarter and then fades. Because prices adjust slowly, the surprise moves output and inflation for a while. The numbers are illustrative: they are not estimates for a country, and the paths are not forecasts.
These values shape how the economy and the central bank react. Their combinations matter, so results can look different from the standard case.
Opening this section also allows persistence up to .
Loading the model…
Prices adjust slowly. Policy matters.
Choose a surprise and run the experiment to see how the economy reacts quarter by quarter.
Every number is a change compared with the path the economy would follow without the surprise, unless a chart says “levels”.
Horizontal axis: quarters after the surprise (quarter 0 is when it happens).
Demand: xt = Etxt+1 − (1/σ)(it − Etπt+1 − rnt)
Inflation: πt = βEtπt+1 + κxt + ut
Central bank rule: it = φππt + φxxt + vt
x is the output gap, π inflation, i the policy rate, and Et means “expected, given what is known in quarter t”.
See how trade and exchange-rate policy change the response to a surprise.
Choose a floating exchange rate with a domestic- or CPI-inflation rule, or a credible fixed exchange rate. Foreign conditions come from a separate model that domestic decisions do not affect.
The numbers are illustrative, not country estimates or forecasts. Domestic inflation measures home-produced goods; CPI also includes imported goods. A rise in the exchange rate means depreciation.
Loading the model…
Choose a surprise and run the experiment.
All values are changes from the path without the surprise. Quantity and exchange-rate percentages are log-percent deviations.
Changes from the path without the surprise. Inflation and interest rates are in basis points per year.
Horizontal axis: quarters after the surprise (quarter 0 is when it happens).
Consumer-price inflation = domestic inflation + import weight × change in the terms of trade.
Interest-rate difference = expected nominal depreciation.
The output gap is relative to natural output, not the Solow steady state. A positive exchange-rate change means depreciation. Log-percent deviations approximate ordinary percentage changes only for small movements. The model continues beyond the last plotted quarter.
See how inflation, interest rates, the output gap, and the exchange rate have actually moved in Canada.
Official Canadian data from Statistics Canada and the Bank of Canada, quarter by quarter. Each line says whether it is observed, a published estimate, or calculated here from observed data.
These are not model results. They show the real-world quantities the model on the Experiments page is about, so you can see how big and how frequent such changes are.
Quarters are three-month periods: 1993Q1 is January to March 1993. Every series is available from 1993Q1; some start earlier.
The output gap is how far actual output is from what the economy can produce without pushing inflation up. It cannot be observed, so the Bank of Canada estimates it in several ways.
Loading the data…
Real data, real quarters
Choose a period to see Canadian inflation, interest rates, the output gap, and the exchange rate.
Each chart covers the same quarters. Hover or tap a point to read its value.
What happens to risk when you put two assets together?
How the expected return and risk of a mix of two assets depend on the share in each, on your attitude to risk, and on how closely the two move together.
The assets and data are made up or simulated for teaching (ECO358 Lecture 5). They are not forecasts or investment advice.
Your Portfolio · Choose two assets
Choose 2 assets
Each point is one asset: further right is riskier, higher up earns more on average. Your two choices are the larger points.
Choose how to split your money, and how much risk you can stand.
·
Your Portfolio · Mix them
The curve shows every share from −150% to 150% in the first asset. Solid: both assets held. Dashed: one is sold short. The dotted curve is your best indifference curve: every point on it is just as good to you.
Your position
Your mix's variance is the sum of these three parts: each asset's own risk, scaled by its share, and how the two move together. A negative covariance part lowers the total.
The two assets keep their expected return and risk; only how they move together changes.
Your Portfolio · Correlation
What do many assets and a safe asset add, and why does everyone end up holding the market?
How many assets form an efficient frontier, how a safe asset turns it into the capital market line, and why the best risky mix must be the market portfolio.
The assets and data are made up or simulated for teaching (ECO358 Lecture 5). They are not forecasts or investment advice.
Choose how many made-up companies to draw, then press Run and watch them arrive.
The Market · Build the frontier
Each dot is a company. Faint lines show the frontier of the first 10, 20, 50… companies; the solid line is the efficient frontier of all of them (no short sales). Any portfolio below the efficient frontier (solid line) earns less for the same risk.
A treasury bill pays a certain return. Draw a line from it and turn it until it just touches the frontier.
The Market · Add a safe asset
Your dotted line is grey while it cuts through the frontier, green when it is the best line, and red when it is out of reach. The solid blue line is the best one; above 100% in M (dashed) you borrow at the risk-free rate.
Three investors, their total, and what happens when one price is wrong.
The Market · Why M is the market
Which risk does the market pay you for?
Why only market risk is rewarded: diversification removes firm-specific risk, beta measures market risk, and a market shock moves each holding by its beta.
The assets and data are made up or simulated for teaching (ECO358 Lecture 5). They are not forecasts or investment advice.
Two of the made-up assets from Your Portfolio, side by side. Nothing to set here: read on, then use the controls further down.
Risk & Beta · Same risk, different reward
Choose how many firms to hold, then press Run: the chart adds them one at a time.
Risk & Beta · Diversification
Solid: the exact risk of holding the first N firms drawn. Dashed: the textbook curve with average firms. Dotted: the market risk no diversification removes.
Pick an asset and a period, then fit a straight line through its monthly returns.
Risk & Beta · Fit a beta
A little history
One dot per month: the market's return (across) and the asset's (up), both above the safe rate. The fitted line's slope is beta.
| Industry | Fitted β | ± 2 s.e. | R² |
|---|
Move the market, or one firm, and see each holding respond.
Risk & Beta · Shocks
How much return should an investment pay for its risk?
The Capital Asset Pricing Model: the return an asset must offer for its market risk, the security market line, how to get the return you want, what happens off the line, and whether the data agree.
The assets and data are made up or simulated for teaching (ECO358 Lecture 5). They are not forecasts or investment advice.
Risk & Beta ended with “investors are paid only for market risk”. The CAPM turns that sentence into a number. Nothing to set here: the controls start in the next section.
CAPM · The model
Why it helps
Move β and read the required return off the line.
Jump to
CAPM · The security market line
How to use the SML
Pick a target. You choose your market exposure; the market sets the safe rate and the premium.
Target: , set above.
Which market condition would deliver it?
CAPM · Your target
The dashed line is your target; where it meets the security market line is the β you need.
Price one asset wrongly, then press Run and let trading push it back to the line.
CAPM · Off the line
What studies of real stock returns found when they drew the line from data.
This section shows an illustration of a documented pattern, not data. Real US data will replace it once their publication is cleared, and its wording will then follow the results.
CAPM · Does it work?
Retire Early · A savings challenge
A game that puts the first four pages to work: build a stock fund and a bond fund, mix them for the best Sharpe ratio, choose how much risk to take, then live through 20 simulated years five at a time.
The securities and data are made up or simulated for teaching (ECO358 Lecture 5). They are not forecasts or investment advice.
Market
Build two funds, mix them, then decide how much of your money to put in the mix.
Retire Early · Your plan
Every mix of your stock fund and bond fund, from all bonds to all stocks. Move the stock share and watch the Sharpe ratio: the expected return above T-bills for each unit of risk.
Retire Early · The years
| Holding | Your plan | Now, after drift |
|---|
Fictional headlines; the numbers are the game's. Each year's change in your wealth, in five parts.
Results appear when the game ends.
Retire Early · The result
| Your plan | The best plan |
|---|