Assessing Economic Complexity with Input-Output Based Measures

Looking at the absolute values of the connectedness measures and their percentage
changes (Table 5) we also note a slight (and perhaps unexpected) reduction in the average
economic complexity, with a decreasing dispersion of countries along the “interrelatedness
scale function” but no significant relative changes, except in the UK, which rose from 8^th in
the 1970s to 4^th in the 1990s.

INSERT TABLE 5

A closer inspection of the absolute values and rankings calls for a careful association of
the measures corresponding to different methodologies or conceptualizations of economic
complexity. This task can be better accomplished by analyzing the correlation coefficients
presented in Tables 6 and 7 and using the following definitions and results.

INSERT TABLES 6 AND 7

Let M be the set of the measures mi, r(i,j) the absolute value of the correlation coefficient
between mi and mj, and c the number 0 ≤ c ≤ 1.

Definition 1: A bundle B of measures of M is a set of elements of M such that, for every
pair (mi , mj ) of B, we have r(i,j) ≥ c and, for every mk of M-B, we have at least one mi of B
such that r(i,k) < c.

Two bundles B1 and B2 are perfectly separated when, for every mk of B1, we have r(k,i)
< c for every mi of B2.

14

<<   12   13   14   15   16   17   18   >>

More intriguing information