*[From the working paper “On the economic importance of the determinants of long-term growth” by Olivier Sterck, posted at the Centre for the Study of African Economies working papers website]*

###### “The long-run growth literature abounds with statements about the“economic signiﬁcance” or “economic importance” of variables of interest. To justify these statements, authors usually report the standardized beta coeﬃcient associated with the variable of interest, or, more rarely, the partial *R*^{2} or the Shapley value.”

^{2}

###### “I show that these statistics are inadequate to measure economic importance.”

###### “I …propose two complementary methods to measure the economic importance of variables and residuals in regressions. … Both methods aim at measuring the percentage contribution of each explanatory variable to variation in the dependent variable. They diﬀer in how they handle variation induced by explanatory variables that are correlated.”

###### “According to the *ceteris paribus* approach, the importance of a variable *x*_{i} is measured as:”

_{i}

###### “The measure *q*_{i}^{2} captures the variation generated by the explanatory variable *x*_{i} *ceteris paribus* and expresses it in percentage terms.”

_{i}

^{2}

_{i}

###### “The second method focuses on *non-ceteris paribus* variation: the importance of an explanatory variable not only depends on the variation it generates alone, but also on the variation co-generated with other explanatory variables. …. According to the *non-ceteris paribus* approach, … the importance of a variable *x*_{i} is measured as:”

_{i}

###### “…I show that it can be interpreted as the elasticity of *Var(y) *with respect to *Var(**β*_{i}x_{i}). The measure is therefore positive if a marginal increase in *β*_{i} increases *Var(y)*, and negative if a marginal increase in *β*_{i} reduces *Var(y).”*

_{i}x

_{i}).

_{i}

_{i}

###### “The two measures are complementary. They bring diﬀerent information, especially when they sharply diﬀer. For example, if the ceteris paribus importance of a variable is much larger than its non-ceteris paribus importance, it means that the eﬀect of that variable is going against the eﬀects of other variables included in the estimated model. On the contrary, if the ceteris paribus importance of the variable is much smaller than its non-ceteris paribus importance, it means that the eﬀect of that variable is reinforcing the eﬀect of other important variables.”

###### “As research analyzing large datasets is more and more frequent, evaluating economic importance is becoming as – if not more – important than assessing statistical signiﬁcance. This research proposed two intuitive methods to measure importance that usefully complement standard measures of statistical signiﬁcance.”

###### To read the article, **click here**.

**click here**