Deriving the Fisher Relation

Why Relate Interest Rates to Inflation?

A nominal interest rate describes how a money balance grows. But a dollar’s purchasing power depends on the prices of goods and services, which may also change during the period. To compare the growth of money with the growth of purchasing power, we need a real interest rate. The Fisher relation connects these rates and inflation, and also explains why the commonly used additive approximation is not exact.

The relation is useful in economic analysis because it separates the change in a money balance from the change in the price level. Its algebra is simple, but its interpretation depends on using the same period for all rates and being clear about whether inflation is treated as known or uncertain.

Definition: In a one-period model, the nominal interest rate \(i\) is the effective annual percentage growth of a money balance. The inflation rate \(\pi\) is the percentage change in the general price index over the same period. The real interest rate \(r\) is the percentage growth in the balance’s purchasing power over that period. When inflation is anticipated, write \(\pi^e\) for expected inflation.

The exact relation is easiest to derive when inflation is known with certainty over the period. In that setting, expected inflation equals the inflation that occurs. The nominal and real rates are both effective rates for the same one-year period, and the price index measures the price of the same basket at the start and end of the period. These assumptions matter: the exact deterministic relation should not be treated as an identity involving only expected inflation when future inflation is uncertain.

Deriving the Exact Fisher Relation

Suppose the price index is \(P_0\) at the beginning of the year and \(P_1\) at the end. If inflation is \(\pi^e\), then \(P_1=P_0(1+\pi^e)\). An initial money balance of one dollar grows to \(1+i\) dollars. To express the growth in purchasing power, compare the amount of goods the balance can buy at the end with the amount it could buy at the start.

1+r = \frac{(1+i)/P_1}{1/P_0} = \frac{1+i}{1+\pi^e}

The numerator of the first fraction is the end-of-period purchasing power of the invested dollar, measured in units of the basket. The denominator is its initial purchasing power. Substituting \(P_1=P_0(1+\pi^e)\) cancels the price index \(P_0\), giving the second expression. Multiplying both sides by \(1+\pi^e\) yields the exact Fisher relation:

1+i=(1+r)(1+\pi^e)

This equation can also be understood as a sequence of growth factors: purchasing power grows by the real factor \(1+r\), while the price level grows by the factor \(1+\pi^e\). The money balance must grow by their product. Expanding the product gives another useful form.

i=r+\pi^e+r\pi^e

The last term, \(r\pi^e\), is the cross-product. It is the reason the exact nominal rate is not generally just the sum of the real rate and inflation. Rearranging the exact relation gives two calculation forms:

i=(1+r)(1+\pi^e)-1 \qquad\text{and}\qquad r=\frac{1+i}{1+\pi^e}-1

Theorem: With a common one-year period and inflation known with certainty, the exact relationship is \(1+i=(1+r)(1+\pi^e)\). Expanding gives \(i=r+\pi^e+r\pi^e\). The additive Fisher approximation \(i\approx r+\pi^e\) omits the cross-product \(r\pi^e\).

Worked Examples

Worked Example: Finding the Nominal Rate

A hypothetical one-year investment is expected to increase purchasing power by 4.0%. The expected inflation rate for the same year is 3.0%, treated as known with certainty. Find the exact effective annual nominal rate and compare it with the additive approximation.

Method. Use the exact relation to multiply the real growth factor by the price-level growth factor. Both rates are annual, so no time-unit conversion is needed.

The real growth factor is \(1+0.040=1.040\), and the price-level growth factor is \(1+0.030=1.030\). Therefore:

$$ 1+i=1.040(1.030)=1.0712 $$

Subtract 1 to obtain the exact nominal rate: \(i=1.0712-1=0.0712\), or 7.12%. The additive approximation is \(4.0\%+3.0\%=7.00\%\). It is lower by \(7.12\%-7.00\%=0.12\) percentage points.

Check. The cross-product is \(0.040(0.030)=0.0012\), or 0.12 percentage points. Adding it to 7.00% gives the exact 7.12%. This check confirms that the difference comes from the omitted product term.

Worked Example: Recovering the Real Rate and Avoiding a Subtraction Trap

A hypothetical account pays an effective annual nominal rate of 10.0%. Expected inflation over the same one-year period is 4.0%, treated as certain. Find the exact real rate, compare it with the approximation, and explain the error in simply subtracting inflation from the nominal rate.

Method. Since the nominal rate is given, divide its growth factor by the inflation factor, then subtract 1:

1+r=\frac{1.100}{1.040}=1.0576923077\ldots

Thus \(r=1.0576923077\ldots-1=0.0576923077\ldots\), or 5.7692% to four decimal places. The additive approximation is \(10.0\%-4.0\%=6.0\%\). It exceeds the exact real rate by about \(6.0000\%-5.7692\%=0.2308\) percentage points.

Common mistake. Writing \(r=i-\pi^e=10.0\%-4.0\%=6.0\%\) and calling this the exact real rate drops the denominator \(1+\pi^e\). The correction is to use \(r=(1+i)/(1+\pi^e)-1\). The subtraction result is only the additive approximation.

Check. Applying the exact relation to the computed real rate gives \(1.0576923077\ldots(1.040)=1.100\), which recovers the original nominal growth factor.

Worked Example: Assessing the Approximation at Different Rates

Compare the exact and approximate nominal rates for two hypothetical one-year settings. In Setting A, the real rate is 2.0% and expected inflation is 1.0%. In Setting B, the real rate is 8.0% and expected inflation is 10.0%. Treat inflation as known with certainty in each setting. For each, calculate the exact nominal rate, the additive approximation, and the discrepancy.

Setting A. The growth factors are \(1.020\) and \(1.010\). Their product is \(1.020(1.010)=1.0302\), so the exact nominal rate is \(1.0302-1=3.02\%\). The approximation is \(2.0\%+1.0\%=3.00\%\). The approximation is lower by 0.02 percentage points, equal to \(0.020(0.010)=0.0002\), or 0.02 percentage points.

Setting B. The growth factors are \(1.080\) and \(1.100\). Their product is \(1.080(1.100)=1.188\), so the exact nominal rate is \(1.188-1=18.8\%\). The approximation is \(8.0\%+10.0\%=18.0\%\). The approximation is lower by 0.8 percentage points, equal to \(0.080(0.100)=0.008\), or 0.8 percentage points.

Conclusion and check. In both settings, the discrepancy equals the cross-product \(r\pi^e\). It is small when both rates are small, as in Setting A, but larger when the rates are larger, as in Setting B. Multiplying the growth factors in each case verifies the exact nominal rates.

When Is the Additive Approximation Useful?

The approximation follows directly from the exact expansion:

i=r+\pi^e+r\pi^e \approx r+\pi^e

It is useful when the cross-product is small relative to the precision needed for the calculation. For example, if both rates are expressed as decimals and are modest, their product will usually be smaller than either rate. But “small” is not a separate rule that guarantees acceptable accuracy: the required precision matters. A difference of a few hundredths of a percentage point may be immaterial in one calculation and important in another.

When solving for a real rate, the approximation \(r\approx i-\pi^e\) is derived by rearranging the additive form. The exact solution is \(r=(1+i)/(1+\pi^e)-1\). For positive inflation and a positive nominal rate greater than inflation, subtracting inflation gives a higher result than the exact calculation. In fact, if \(r\) is the exact real rate, then the difference between the approximate rate \(i-\pi^e\) and the exact rate is \(r\pi^e\).

(i-\pi^e)-r = (r+\pi^e+r\pi^e-\pi^e)-r = r\pi^e

This comparison uses the exact relation and the stated sign conditions. If rates are negative or inflation is negative, do not infer the direction of the approximation error from the positive-rate case; use the exact formula or check the cross-product directly.

Assumptions and Common Exam Traps

The exact relation is an effective-rate formula for matching periods. If a rate is quoted for a different compounding frequency, first express it as the effective rate for the period being analyzed. This tutorial uses effective annual rates, so the nominal rate \(i\), real rate \(r\), and inflation rate \(\pi^e\) all refer to one year. Combining an annual interest rate with monthly inflation, for example, would not produce a meaningful direct comparison.

The certainty assumption is also important. If future inflation is uncertain, realized inflation can differ from expected inflation, and a single expected inflation number does not by itself make the deterministic purchasing-power derivation exact. The exact equation above applies directly when the inflation rate for the period is treated as known. In the usual approximate Fisher statement under uncertainty, expected inflation is used to summarize anticipated price growth; the additive relation remains an approximation rather than a guaranteed identity.

Keep these checks in view when using the relation:

Match the period. Use rates defined over the same interval and on a consistent effective-rate basis.
Preserve the growth factors. The exact equation multiplies \(1+r\) by \(1+\pi^e\); it does not multiply the rates \(r\) and \(\pi^e\) alone.
Distinguish exact from approximate. The exact nominal rate includes \(r\pi^e\); the sum \(r+\pi^e\) does not.
Use the right rearrangement. To find a real rate from a nominal rate, divide growth factors before subtracting 1.
Check the assumption. An expected inflation rate is not automatically a known inflation rate.

Key takeaway: Under a one-period certainty model, nominal money growth equals real purchasing-power growth multiplied by price-level growth: \(1+i=(1+r)(1+\pi^e)\). The familiar sum \(i\approx r+\pi^e\) omits the cross-product \(r\pi^e\), so use the exact relation when precision or larger rates make that term material.

Check Your Understanding

For each problem, use effective annual rates over the stated one-year period. Unless a problem says otherwise, treat expected inflation as known with certainty.

A hypothetical savings product has a real rate of 3.0% and expected inflation of 4.0%. Find its exact nominal rate and the additive approximation. (Answer: Exact 7.12%; approximation 7.00%.)
A hypothetical investment pays a nominal rate of 7.0% while expected inflation is 2.0%. Find the exact real rate and compare it with the subtraction approximation. (Answer: Exact real rate about 4.9020%; approximation 5.0%.)
A one-year contract has a nominal rate of 9.0% and expected inflation of 5.0%. Calculate the exact real rate, the approximate real rate, and their difference in percentage points. (Answer: Exact real rate about 3.8095%; approximation 4.0%; difference about 0.1905 percentage points.)
Compare the exact and approximate nominal rates when the real rate is 2.0% and expected inflation is 8.0%. State the discrepancy and identify the omitted term. (Answer: Exact 10.16%; approximation 10.00%; discrepancy 0.16 percentage points, the cross-product \(0.020(0.080)\).)
An analyst uses the exact certainty formula with expected annual inflation of 3.0%, but the nominal rate provided is an effective monthly rate of 0.5%. Explain the unit problem and state what must be done before applying the annual Fisher relation. (Answer: The periods do not match; convert the effective monthly rate to an effective annual rate, \((1.005)^{12}-1\), before using the annual relation.)

What You'll Learn