# Published in IZA World of Labor

### “The Mincer equation gives comparable estimates of the average monetary returns of one additional year of education

Elevator pitch

The Mincer equation—arguably the most widely used in empirical work—can be used to explain a host of economic, and even non-economic, phenomena. One such application involves explaining (and estimating) employment earnings as a function of schooling and labor market experience. The Mincer equation provides estimates of the average monetary returns of one additional year of education. This information is important for policymakers who must decide on education spending, prioritization of schooling levels, and education financing programs such as student loans.

## Key findings

**Pros**

- Earnings can be explained as a function of schooling and labor market experience using the Mincer equation; this provides policymakers with important information about how to invest in education.
- Due to the comparability of Mincerian results, individuals can make use of these results to help guide their personal decisions about how much schooling they should invest in.
- Recent studies using the Mincer equation indicate that tertiary education, as opposed to primary education, may now provide the greatest returns to schooling; this represents a shift in the conventional wisdom.

**Cons**

- The relationship between schooling and earnings does not necessarily imply causality.
- Earnings functions provide private (i.e. individual) returns to schooling, whereas government/public costs and other benefits are needed to estimate social rates of return.
- As economies become more complex and technological developments alter the demand for education, decades-old cross-sectional data may not be informative about returns to current investment decisions.

## Author’s main message

The Mincer equation suggests that each additional year of education produces a private (i.e. individual) rate of return to schooling of about 5–8% per year, ranging from a low of 1% to more than 20% in some countries. Globally, the returns to tertiary education are highest, followed by primary and then secondary schooling; this represents a significant reversal from many studies’ prior results. Policymakers can learn much from Mincerian results; for instance, further expansion of university education appears to be very worthwhile for the individual, meaning that governments need to find ways to make financing more readily available, and that high rates of return are found through investment in girls’ education.”

Very interesting analysis.

Personally, I am interested in Mincer equation, but I have some questions.

What about small values of R square of the model/equation. And how do we calculate/interpret the monetary returns of one additional year of experience?

Many thanks in advance.

LikeLike

Olga, a small R-square is not the problem. In fact, you can have an equation with an R-square less than .3 and you are ok. If a Mincer equation has such an R-square, then you are showing that 2 variables — education and experience — are explaining about a third of the variation in earnings.

I would not over-interpret the coefficient on experience. It is in the equation to model the evolution of earnings over the life cycle.

Thanks

Harry

LikeLike

Harry, many thanks for your reply.

I suppose that having an extended form (with other variables, e.g. gender or ethnicity) of Mincer equation and with such an R-square the explanation is the same, the amount of variables Var1,Var2,Var3…, Varν, which are significant explaining about a third of the variation in earnings.

Specifically, I am thinking of a Mincer equation that has the following form:

log (Y) = log (yo) + bS + cX + dX^2 + ε, where, S: years of education, X: years of (potential) experience and ε: all other variables.

But then I have to be careful with the returns of the independent variables, because my dependent variable is expressed on a logarithmic form, correct?

Do you apply the following?

• Increase of one-unit of variable S (years of education) with coefficient b, then the change for dependent variable is e^b ?

• For small values of b (b < |0.1|) και e^b ≈ 1 + b, then the change for dependent variable is (100 × b)?

Many thanks,

Olga

LikeLike

Is it possible please to answer on the above questions?

I would be very grateful, because I am not sure on the explanations.

Many thanks in advance,

Olga

LikeLike

Olga,

Halvorsen and Palmquist (1980) said that the coefficient cneeds to be transformed by exp(c)-1 for it to be interpreted as the schooling premium. But then Kennedy (1981) pointed the value of c with certainty is needed and said that the estimated coefficient c’ needs to be transformed by exp[c’–½V(c’)]–1 before it can be interpreted as a premium, where V(c’) is the variance of the estimated coefficient c’.

When the values are small, you are correct.

Kennedy, P. (1981) “Estimation with Correctly Interpreted Dummy Variables in Semilogarithmic Equations,” American Economic Review, 70:1.

Halvorsen, R. and Palmquist, R. (1980) “The Interpretation of Dummy Variables in Semilogarithmic Equations,” American Economic Review, 70:3.

Cheers

Harry

LikeLike