Long-term constraints and economic objectives (Study)

Abstract:

· Four apparent long-term dynamics may require new economic responses from developed countries and prompt us to rethink economic objectives for these countries;

· The first: a decline in total factor productivity growth in developed countries;

· The second: technological change that appears to generate little growth and could increase inequality;

· The third: demographics, particularly the aging of the population;

· The fourth: environmental constraints.

Any genuine economic policy must first clearly define its objectives and thus require reflection on social objectives that include ethical, moral, philosophical, and material elements. This is because the economy is not an end in itself, but a means to an end.

However, economic constraints shed light on possible economic objectives and the trade-offs to be made, and are therefore an integral (but of course not the only) part of this process of defining our social objectives and, consequently, the economic objectives for achieving them. We are currently facing four long-term phenomena that must be taken into account when determining our economic and social objectives.

The findings

Finding 1: Decline in TFP growth

The first finding is based on the observation of a decline in total factor productivity (TFP)[1] growth in developed economies. Economist Robert Gordon (2014, 2015a, 2015b) argues that the third industrial revolution (electronics and digital technology) probably has much lower productivity growth potential than the second industrial revolution (electricity, the internal combustion engine, the telephone, radio, chemical engineering, advances in the fight against infectious diseases, etc.)[2].

He shows that the annual growth rate of TFP in the United States, which was 2.17% from 1920 to 1950 and then 1.79% from 1950 to 1972, declined to 0.52% and 0.54% in the periods 1972-1996 and 2004-2014. Between 1996 and 2004, TFP growth rebounded to 1.43% due to new technologies, but he believes this phenomenon is unlikely to be repeated. Put simply, new technologies have made it possible to speed up routine tasks in the service sector with the widespread use of microcomputers and to improve the efficiency of supply chains, particularly in distribution. This has significantly increased productivity.

However, at least in the United States, this phase is largely over. The second industrial revolution—which was particularly long and gave the majority of people access to drinking water, electricity, and cars—came to a halt in a sense between 1972 and 1996. At that point, progress came from smaller improvements in terms of raising living standards. Machines and changes in production processes were in a much less revolutionary phase than those seen in the era of Henry Ford, for example. The digital revolution seems to have reinvigorated the potential for innovation, but only for a short time, with a decline already visible in the creation of start-ups and young, fast-growing companies.

Observation 2: Technology that generates little growth and creates inequality

To quote Gordon (2015b): « This industry, which employs all these engineers and owns all these factories and salespeople, needs you to throw away your old stuff and buy new stuff—even if the new stuff is only slightly better. » While the hypothesis of low productivity growth remains open to debate, it is nonetheless plausible for developed countries, as it is likely to limit per capita GDP growth. We must therefore take this perspective into account when setting our objectives.

New technologies, which do not guarantee strong productivity growth, can generate inequality. First, we are seeing the destruction of mid-skilled jobs as a result of new technologies[3], which tends to erode the incomes of part of the middle class[4], alongside an increase in the premium for high skills.

· In France, the correlation between PISA reading test scores and a socio-economic-cultural origin index is the highest of all OECD countries[5]. With a high premium on qualifications, this correlation can not only maintain intergenerational inequalities but even accentuate them.

· New technologies can also create inequalities because they tend to be characterized by a « winner takes all » effect. This leads to monopolies and therefore rents, which can generate inefficiency and reinforce the power of the richest through political processes[6].

· Finally, marked inequality can potentially lead to lower production levels through lower aggregate demand.

Observation 3: Population aging

The aging population will have a negative impact on per capita GDP growth. This topic was discussed in detail in one of our previous articles[7]. The logic can be summarized by the fact that aging reduces the ratio of the working-age population to the total population, which automatically reduces GDP per capita in the absence of an increase in productivity, an increase in the employment rate of the working-age population, and/or an increase in the number of hours worked per person employed.

As noted in Gordon (2014), other demographic factors can also have a negative impact on per capita GDP growth. Indeed, the high average growth rates during the 20th century were also the result of higher female participation rates, made possible in particular by technological progress and social changes, and the democratization of education. The first factor increases the ratio of workers to total population and, in particular, helped maintain high GDP per capita growth in the United States during the period 1950-1972, despite the slowdown in productivity growth. The second factor increases the skills and therefore the productivity of the workforce. It is certain that the first factor will have less of an impact on future growth. As for the second factor, its positive impact is likely to be less significant. As with technology, the idea of diminishing returns applies here.

Observation 4: Environmental constraints

Finally, environmental constraints will inevitably come into play due to the issue of global warming and limited resources for a world population that is expected to reach nine billion by 2050, with a higher standard of living in developing countries and therefore potentially higher demand. In particular, on the energy front, replacing diesel cars with electric cars can be seen as a source of progress, particularly because of the reduction in negative impacts on the environment. But ultimately, we are replacing one energy source with another. It was not the « replacement » of the horse by the combustion engine that increased the available energy tenfold and enabled the spectacular progress we know today. In terms of living standards, there are gains to be made from the reduction in negative externalities (pollution), but these are unlikely to materialize in material gains measured by GDP.

What are the consequences?

The factors presented above suggest rather weak future growth in GDP per capita. First of all, we can put this into perspective. Low growth for Haiti necessarily implies that the standard of living there will remain dramatically low (in 2014, GDP per capita in purchasing power parity was USD 1,731.8 compared to USD 38,847.5 for France, according to the World Bank).

For a developed country such as France or the United States, low GDP per capita growth simply means that the average standard of living will remain high and will increase, but only slightly, from year to year. This is by no means a disaster, at least not from an economic point of view. It should be noted, for example, that during the thirty glorious years, often presented to us as a golden age, GDP per capita was much lower than it is today. In 1970, it was about half what it is today (see graph below).

Figure 1 – Change in GDP per capita in France, chained prices in euros, base 2010 (sources: INSEE, BSI Economics)

Of course, we might wish for a higher standard of living, and there are ways to achieve this: a more effective education system for France, a better match between training and the skills required in the labor market, and a more efficient job market with training support to increase the employment rate. And it is undoubtedly possible to improve various aspects of material life without increasing ecological damage, even if the presence of the latter will most likely lead us to desire lower per capita GDP growth for some of these components[8].

On the other hand, low per capita GDP growth raises certain questions:

· The first is that of inequality, if we believe that the main problem is not today’s average level but its distribution. In an environment of strong growth, the question of distribution is less problematic from a political point of view. Indeed, it is easier to increase redistribution when the average standard of living rises, as it is theoretically possible to improve the situation of poorer households without worsening that of the richest. In the extreme case of zero growth, on the contrary, the standard of living of the richest households must necessarily be reduced in order to increase that of the poorest.

A second factor is unemployment. If we believe, for example, that work is a vehicle for integration and social belonging, then income redistribution—even aside from its incentive effects—is not the only social objective, and a low unemployment rate can be an objective regardless of individual income levels. The problem is that, until now, reductions in unemployment have been linked to high growth rates. The question of « how can we reorganize our labor market so that it generates low unemployment even with low growth? » therefore arises insistently if low unemployment is an objective in itself. It may also lead to changes in our vision of social organization, particularly since the latter is an essential component of well-being beyond material aspects. This brings us back to the fact that GDP is a very imperfect and limited measure of well-being, as it is highly materialistic.

A third issue is that of pensions. If the consumption patterns of older people were similar to those of young people, the problem of pensions would be the same as that of GDP per capita. As we have seen, in this case, aging tends to lower GDP per capita. If we want to maintain the same level of GDP per capita, other components of GDP must offset the negative effect of aging. However, the problem is more complex because the demand and savings patterns of older people are very different from those of young people. In particular, their need for labor-intensive (low-productivity) personal and medical services, which are unlikely to see significant productivity growth, poses a more significant problem.

If retirees need a significant amount of low-productivity services to maintain a satisfactory standard of living and their numbers increase sharply in proportion, then this implies that, in order to maintain this standard of living, a significant number of workers will have to be reallocated from the productive sector to the low-productivity sector. But this will have a significant negative effect on the production of other types of goods and therefore on the quantity that can be consumed. We can therefore expect a negative effect on the standard of living of the working population if, for example, productivity growth in at least one of the two sectors is not strong enough. In other words, demand that is shifting towards low-productivity goods is demand that costs more in terms of labor (which is, moreover, the resource that is becoming scarce in aging economies).

As a result, the real challenge for growth, in order to maintain a certain standard of living, may lie in these low-productivity sectors, where demand is expected to grow strongly. In this case, it is undoubtedly more difficult to define, but it involves better adaptation of housing and the organization of living spaces. It is also clear that the above discussion reinforces the argument in favor of preventive measures.

Finally, and this is a key point, if productivity in these services increases only slightly, a large number of people will need to be trained to perform them. Some require high qualifications, which must be anticipated because training can be lengthy. Overestimating productivity growth in these sectors could lead to insufficient training, with potentially very negative consequences for well-being.

Conclusion

Low growth—reflected in particular by low TFP growth—raises the issue of inequality, especially when technological change tends to have a negative impact on the middle class. It may also prompt us to rethink our labor market, or our social organization, in a more radical way if we can no longer count on strong growth to reduce unemployment.

Furthermore, aging leads to a reallocation of demand towards sectors with low productivity, and therefore labor-intensive, which, in addition to calling for measures to improve productivity and workforce training, risks reopening the debate on living standards between generations. Finally, environmental constraints are forcing us to rethink material progress in a different way.

References

Antonin Bergeaud, Gilbert Cette, and Rémy Lecat, Productivity Trends from 1890 to 2012 in Advanced Countries, Banque de France Working Paper No. 475, February 2014.

Gilbert Cette, Productivity growth: what prospects for France, September 26, 2013.

Daniel Cohen, The World is Closed and Desire is Infinite, Albin Michel, 2015.

Célia Garcia-Peñalosa, Patrick Artus, and Pierre Monhen, Restoring France’s Potential Growth, CAE Note No. 16, 2014.

Robert Gordon, The Demise of U.S. Economic Growth: Restatement, Rebuttal, and Reflections, Working Paper, January 2014.

Robert Gordon, Secular Stagnation: A Supply-Side View, American Economic Review: Papers & Proceedings, 105(5):54-59, 2015a.

Robert Gordon, Secular Stagnation on the Supply Side: U.S. Productivity Growth in the Long-Run, Digiworld Economic Journal, 2015b.

OECD, Overview of Structural Reform Actions in the Policy Areas Identified as Priorities for Growth, Going for Growth Interim Report, 2014.

Joseph Stiglitz, La Grande Fracture, Les Liens qui Libèrent, 2015.