Why Virtuous Supply-Side Effects and Irrelevant Keynesian Effects are not Foregone Conclusions: What we Learn from an Industry-Level Analysis of Infrastructure Investments in Portugal

We use the industry-specific effects of twelve different infrastructure investments in Portugal to inform about the mechanisms through which such investments affect economic activity. Our main findings are as follows. First, demand-side effects that are approximated by adding the short-term and long-term construction effects are very important. They are over 60% of total effects for airport investments, ports, refineries, and water, and over 45% for national roads, municipal roads, telecommunications, health and education. Second, site-location effects that are approximated by real-estate effects are also very significant, in particular for national roads, highways and railroads, with 30%, 35% and 64% of the total effects, respectively. They are negative for water and electricity, and zero for municipal roads, airports, and refineries, and negligible for ports, i.e., all these are cases in which we would expect adverse or small location effects. Third, the functional channel relating to internationally-traded goods, approximated by the effects on the primary sector, on manufacturing, and on transportation, is much less significant, although we estimate meaningful effects on heavy industry from investments in all types of road infrastructures, ports, health, and education, as well as on light industry from ports. Fourth, a functional effect affecting non-traded industries, mostly private and public services is much more relevant. It accounts for more than 30% of the effects in the cases of municipal roads, airports, and refineries, and in excess of 20% for highways, railroads, telecommunications, health and education. The fact that most functional effects accrue to non-traded industries is likely to affect international competitiveness adversely. Naturally, these results cannot be automatically generalized, as the nature of the effects of infrastructure investments crucially depends on the level of development of the country in question, on the maturity of its existent infrastructure systems, and on the rigor of all decisions pertaining to infrastructure investment. Nevertheless, they establish that, as infrastructure investments are concerned, the dominance of virtuous supply side effects is not a foregone conclusion and, conversely, the relevance of Keynesian effects cannot be dismissed.