Assessing Economic Complexity with Input-Output Based Measures

Another, more useful distinction is obtained by considering, within the domain of
strongly correlated non-Boolean measures, the correlation coefficients with positive and
negative signs, pointing to a further separation of bundles of this kind:

B11 = {PINT, AVOM, MICT, IDET, DEVA, MPLE, APLU, CAIL}

B12 = {CYCI, CAIA}

The closely related behavior of measures CYCI and CAIA is supposedly explained by the
fact that they explicitly exclude (direct) intra-dependence flows (the values of self-supplying
inputs or the coefficients in the main diagonal of matrix A), and that, in the sense of
complexity as (sectoral) interdependence, these are probably the most appropriate measures.

To explore this distinction further, it is useful to calculate the rankings of economic
complexity based on bundle B11 (Tables 8 and 9) and bundle B12 (Tables 10 and 11).

INSERT TABLES 8, 9, 10 AND 11

The interesting result is, of course, that, according to this particular notion of complexity
as interdependence, large economies appear to be less complex than small ones and
complexity does not necessarily increase as economies grow and develop. But the full
understanding of all the forces behind this surprising result would require further research.

16

<<   14   15   16   17   18   19   20   >>

More intriguing information