Should you buy or rent?
Buying a home can be a good financial investment, but has high up-front and on-going costs. Renting has lower costs, letting you invest more capital elsewhere.
How it works
The model runs by simulating 1,000 "possible futures" based on the parameters you specify. In Causal, model parameters are probabilistic — they can take on a range of values to account for the uncertainty in your estimates. Causal's outputs are distributions too, letting you understand the uncertainty in your results.
Customise the inputs to run your own scenarios comparing buying and renting.
Inputs
Model
- Type: Random
- Expected Value: 5.0%
- 95% Confidence Interval: 3.6% – 6.4%
Outputs
- Rent
- Buy
For renting, this is the initial investment in stocks.
Spending
Wealth
If you know Excel, you'll quickly pick up Causal.
The Model
We compare the two scenarios by considering, at different points in time, the wealth accumulated by each. The interesting time point is when one scenario "overtakes" the other — this answers the question "If I'm planning to stay in a property for X years, is it better to buy or to rent?".
Here's how each scenario was modeled. Example figures are given in square brackets.
Scenario 1: Buying
When buying a property of a certain Property Price [$500,000], you need some initial capital:
- Down-payment [$100,000] — a lump-sum that you pay up-front (typically 20% of the property price)
- One-off costs [$20,000] — various fees involved in the property purchase (typically 3-6% of the property value)
You then take out a loan with a particular Loan Term [30 years], for the property price less the down-payment [$400,000]. The Interest Rate [3–5%] of your loan fluctuates slightly each year, and each year, you pay off part of the loan, and pay interest on the remaining unpaid part.
Each year, you pay Annual Costs [2%] for property maintenance, and each year, the value of your property increases according to the rate of Annual Appreciation [2–7%].
Each year, the wealth you've accumulated is equal to the current value of the property minus the loan amount outstanding. In practice, there will be costs involved with selling the property that our model doesn't account for. Property is also relatively illiquid — there's no guarantee that you'll be able to sell the property at your desired price and on your desired timeline.
Scenario 2: Renting
When renting, there are no up-front costs. You make an up-front investment into the stock market, of the same size as the Down-payment and the One-off costs, which earns you annual Market Returns [4–10%].
Each month, you pay Monthly Rent [$1k]. We compare this figure to the equivalent monthly mortgage payment under the Buying scenario. If the monthly rent is lower, the difference is invested into the stock market under the Renting scenario. If the monthly rent is higher, the difference is invested into the stock market under the Buying scenario.
Each year, we assume that your rental payments increase according to the rate of Annual Appreciation.
Each year, the wealth you've accumulated is equal to the current value of your investments. Stocks are relatively liquid, so you'll likely be able to sell them at "market value" at any time.
When analysing a real-world scenario, there is always a trade-off between model complexity and accuracy. For simplicity, we've ommitted some real-world factors, like tax incentives and deductions here.
Technical Details
Probabilistic modeling
Unlike spreadsheets, nodes in Causal can be probabilistic, to account for uncertainty in estimates.
It's hard to say exactly what Market Returns will be over the next 10, 20, 30 years. But we can reasonably say that they'll probably be 4–10%, and that there's a small chance they may even be significantly lower or higher. Representing variables as probability distributions lets us do this.
Simulation
Causal's models run via simulation. By sampling appropriately from each probabilistic node, the Monte Carlo engine simulates 1,000 "possible futures". So instead of a single output number, Causal outputs a distribution representing the range of plausible outcomes.
What's wrong with spreadsheets?
Spreadsheets can't handle uncertainty.
Most quantities in a model aren't precisely known, but spreadsheet cells hold fixed values. Without uncertainty, you can't know how confident to be in your results.
Causal’s models are probabilistic, letting you understand the uncertainty of your results.
Spreadsheets aren't collaborative.
Spreadsheet models are opaque — it's hard for your team to understand your model, even after tracing through each cell and its references one by one.
Causal’s visual diagrams and abstractions make it clear how your model works.
Spreadsheets can't connect to data.
To work with datasets, you have to pass around CSVs in email attachments, or build data pipelines to periodically export data from other sources.
Causal has live data integrations with SQL databases, APIs, and your existing SaaS products, like Stripe and Salesforce.
If this resonates with you, please leave your email above, or get in touch: hi@causal.app.
We'd love to hear from you.