What Will Inflation Be? What Will Happen to the Prime Rate? How Will the Variable Rate Change?

Have you decided to build a mortgage mixture that includes variable-rate, prime-linked, or CPI-linked tracks? Then you have to make assumptions about where these indices are headed. You have to take a shot at answering "what will inflation be" and "what will happen to the prime rate." If you don't, you're defaulting to the "it'll be fine" approach. And surely we can agree that when it's our home and our family's home on the line, we should take a more responsible approach.

Although there's a standard way today to model changes in these economic indices, we still think it's worth walking through the wrong approaches we've run into over the years.

First wrong approach: "What was, will be"

What it says: We'll assume that inflation and the prime rate in the coming years will be whatever they were recently. "What will inflation be over the coming years? Whatever it was last year."

Why we think this isn't good enough: The mortgage market is dynamic. Some years the interest rate is very low, others it isn't. Inflation could be low this year and very high the next. This approach is especially shaky in economically very good or very bad years.

We're better off taking a more flexible approach that asks: what happens if what was isn't what will be?

• In a period of bad conditions, we might over-hedge our mortgage and end up paying for a more expensive one.
• In a period of good conditions, we might choose a riskier mortgage that turns dangerous once the macroeconomic situation sours.

Second wrong approach: "One size fits all"

What it says: We'll plan a mortgage mixture using a single fixed number for the economic variables. "What will inflation be over the coming years? 2%!"

What this means in practice: One single number applies to inflation, to the prime rate, and to the variable rates for the next thirty years. For example, inflation is set at 2% a year — because that's the midpoint of the Bank of Israel's inflation target. The prime rate is set at 3%.

Why we think this isn't good enough: First off, this approach is much better than the previous one. As long as the constants you pick aren't too low, it hedges the risks better.

The first problem is: which number do you use? What do you plug in to most reasonably capture what will happen over the next thirty years? For inflation it's easy, since we know the Bank of Israel wants annual inflation between 1% and 3%. But what do you put for the prime rate? And for the variable rates? The economic climate is always shifting, and pinning a single number on thirty years of Bank of Israel interest rate changes strikes us as crazy.

These are the main approaches we've run into over the years. We call them "problematic" because they rest on rules of thumb and unsubstantiated assumptions. The approaches we'll describe next — the ones we actually use — are grounded in academic research, and what's more, they're used by governments, central banks, and financial institutions.