The Cecchini Report estimated that, by making the allocation of resources within the EC more efficient, 1992 would produce a 2.5-6.5% increase in the level of EC output. But 1992 could have even more dramatic effects on real standards of living, Richard Baldwin told a CEPR lunchtime meeting on 8 November, if one takes into account the impetus these one-off efficiency gains will give to investment and growth.
Richard Baldwin is a Research Fellow in CEPR's International Trade programme. His talk was based on pioneering research reported in `The Growth Effects of 1992', published in November in Economic Policy No. 9, the first of two special issues of the journal devoted to 1992. Issue No. 9 is available for £10.50 from The Journals Publicity Department, Cambridge University Press, The Edinburgh Building, Shaftesbury Road, Cambridge, CB2 1BR, tel. 0223 325807. Financial support for the meeting at which he spoke was provided by the Economic and Social Research Council, as part of its support for the Centre's dissemination programme. The opinions expressed by Baldwin were his alone, however, and not those of the ESRC or of CEPR, which takes no institutional policy positions.
1992 aims to remove all barriers to the free movement of goods, people and capital in the EC and to eliminate distortions that prevent efficient allocation of resources, such as cross-country variations in indirect taxation and regulation. The Cecchini Report estimated that these efficiency gains would lead to a one-off rise in total EC income of 2.5-6.5%, fully realized after 5-7 years. Though this is not negligible, Baldwin noted, it hardly matches up to the radical nature of the programme. But there are good reasons to expect the improvements in resource allocation after 1992 to be at the lower end of the Cecchini range, which is much higher than consensus estimates of the impact of other major liberalizations, such as the Kennedy and Tokyo Round tariff cuts or the US-Canada Free Trade Agreement.
These apparently conflicting responses can be reconciled, according to Baldwin: 1992 will not only bring a one-off jump in the level of EC income, it will also have dynamic effects on the rate of output growth, which the Cecchini Report did not estimate. The efficiency gains from improved resource allocation will mean that Europe can get more output per input of labour and capital. As the returns to capital increase, so the investment climate in Europe will improve. The resulting rise in investment will increase the EC's capital stock and thereby create a knock-on effect on the level of GDP.

The keys to quantifying this `medium-run growth bonus' are the relationships among investment, productivity and output, and in particular the capital-output elasticity the increase in output stemming from a 1% increase in the stock of capital. It is very difficult to estimate these relationships. Baldwin had specified a relatively simple model of investment, productivity and output, taking estimates of capital-output elasticities for several EC countries from the work of other researchers. He `calibrated' his model by finding values for the remaining parameters that enabled it most closely to reproduce observed data. Calibration techniques are far from ideal because they cannot be subjected to conventional tests of precision, but they represent the best way of dealing with data inadequacies in exercises such as this.
The growth effects of 1992 will be proportional to the size of the one-time gains. Baldwin's estimates suggested that they could be large, 40% to 200% over the one-time effects. So, taking the upper and lower bound of Cecchini's estimates, the one-time and growth effects of 1992 together could eventually add anything between 3.5% and 19.5% to Europe's real standard of living, about half of it within 10-15 years. This implied a rise in the rate of EC growth of around 0.2 percentage points a year over this period. After that the effect would fall off due to diminishing returns to capital. The calculations are in many ways quite exploratory, Baldwin cautioned, so the actual numbers should be taken with a grain of salt. But the conclusion that the growth effects of 1992 will be significantly greater than the one-time estimates produced by Cecchini is robust; eventually even very small growth effects will far outweigh one-time gains.
The assumptions underlying Baldwin's calculations are taken from the conventional, `neo-classical' view of growth. But economists such as Paul Romer have recently challenged the validity of the neo-classical model, claiming that it provides no convincing explanation of why Japan grows faster than the United Kingdom or of why growth slowed down worldwide in the 1970s. This has led to the development of the `new growth economics', which emphasizes economy-wide increasing returns to scale.
The hypothesis of economy-wide increasing returns to scale is based on the intuition that investments in R&D depend on the size of the market faced by firms. As demand rises, a higher marginal return to capital increases the returns to innovation; as more is invested in R&D, the process of innovation itself accelerates. Baldwin had conducted econometric tests of this hypothesis using historical data on capital per worker and productivity for the six major industrial countries in the period 1880-1980. Though the results were not conclusive, neither they nor other estimates suggest that the new growth economics should be dismissed out of hand, he maintained.
Economy-wide increasing returns to scale would mean that a one-off increase in EC economic efficiency and output could have yet further beneficial effects. Unlike the `medium-run growth bonus', which will eventually taper off, economy-wide increasing returns to scale would translate the one-off gains into a permanent, sustainable increase in EC growth. Further calibration exercises conducted by Baldwin using two such new growth models suggested that 1992 might permanently add 0.25 to 1 percentage point to the annual rate of EC output growth. Though highly controversial, if these numbers are even close to correct, the impact of 1992 on Europe's standard of living could eventually be enormous.