The unknown potential of the Dutch economy
To the Outlook 2017 overview page
- Potential growth is important for the future prosperity of the Netherlands
- Potential growth is determined by labour, capital and technology
- The growth of the labour supply is stagnating due to the ageing of the population
- The rate of capital deepening remains on track
- Higher growth has to come from further technological progress
The crisis is behind us, and the Dutch economy is again posting average growth of two per cent a year. Forecasts of permanently lower economic growth can apparently be shelved. But the future growth forecasts are surrounded by many uncertainties. This is why we need a more detailed analysis of the potential for growth. And indeed, it is not at all certain that growth of two per cent can be maintained for a longer period.
Potential growth is determined by the supply side of the economy
In the long term, economic development is driven by technological progress, labour and capital. These factors form what is known as the supply side of the economy, and determine what the economy can potentially produce. If we want to estimate the growth potential of the Netherlands for the coming period, we need to make a good analysis of the supply side (the development of technology, labour and capital). Of course, supply and demand do not perfectly match each other at all times. Actual growth in the short and medium to longer term therefore is often not the same as potential growth. We discuss this possibility in our article on the demand side of the economy.
Potential growth can be divided into the contributions from the labour supply and labour productivity. The more people looking for work in the labour market, the higher the potential growth. And the more capital that is committed, or the more rapid technological progress is, the more productive they become.
This article accordingly seeks to estimate the most likely development of the factors of labour and labour productivity - a base scenario. The calculations, which we explain below, produce an average contribution from structural employment growth of 0.2 per cent per year for the 2018-2023 period and an average contribution from structural labour productivity growth of 1.0 per cent per year. The average potential economic growth thus amounts to 1.2 per cent per year for the 2018-2023 period (table 1).
Obviously, the underlying determining factors may not turn out as expected. It is therefore possible that technological breakthroughs could generate a significantly more rapid growth rate in the coming years. We have developed a separate scenario for this specific situation.
Development of labour supply in the coming years
Labour will make less of a contribution to potential growth in the coming years than was previously the case. This has everything to do with demographic developments. As in many Western countries, the Dutch population is ageing, so that the growth of the potential working population is flattening off and will actually decline over time (after 2023). The average age of people in work is also rising. Older people generally participate less frequently in the labour process, so that the average degree of participation by the society is declining. This is known as the cohort effect. On the other hand, the increase in the retirement age has to some extent addressed the ageing problem and has increased the size of the potential working population.
In order to estimate the net growth of the labour supply in the coming years, we calculated structural degrees of participation per age group and combined these with data on the size of each population group.
Figure 1 shows the resulting development of the working population (all people in work and those looking for work aged between 15 years and retirement age), including the effects of a later age of entitlement to state retirement pension (AOW). The number of people looking for work and (in a situation of equilibrium) actually finding it will rise by 0.2 per cent per year on average in the 2018-2023 period, whereby the labour supply will decline slightly at the end of this period.
Due to the effects of ageing, the growth of the labour supply will thus be permanently lower than before the crisis. For instance, in 2002-2008 this growth was still at a rate of 1.0 per cent. The increase in the labour supply is therefore expected to have only a limited effect on growth.To achieve a higher growth rate, labour productivity has to increase as a result of extra commitment of capital, better educated labour or technological progress.
Capital looks as though it will continue to add value to potential growth in the coming period. Capital investment declined temporarily during the crisis due to a loss of confidence and cyclical effects, so that the rate of capital deepening fell: investment in machinery and automated processes per employee declined. In addition, there was a situation of labour hoarding, with businesses retaining employees rather than firing them (CPB, 2011). Both these factors lowered the growth of labour productivity.
Whether these effects are negative in the longer term is debatable. Lower investment at macro level does not necessarily mean that investment in the stock of capital goods declines. And even in a situation of low growth, it can still be very profitable for businesses to invest if the investment leads to productivity gains. The effect of low growth on investment (Bhageloe, 2016) does not necessarily lead to lower potential growth.
The data do not indicate any permanent negative effects from the crisis. Looking at how the trend of capital deepening is developing, we estimate the average annual contribution to growth will be 0.4 per cent in the 2018-2023 period.
The last factor relevant for potential growth, the growth of total factor productivity (TFP), stagnated during the crisis and does not as yet look as though it is returning to its previous growth trend. TFP growth is the purest measure of the status of technology. It shows the efficiency with which capital and labour add value.
According to this, the technological progress that we see around us will make only a limited contribution to higher growth. How can it be that today’s innovations such as smartphones, biotechnology, artificial intelligence, 3D printing and big data are not having a greater impact?
This question has led to an intense debate among economists. One explanation is that while these innovations significantly affect productivity, their effect is not properly measured (Mokyr, 2015). For example, you can book a holiday using mostly free apps and platforms rather than a travel agency, but this economic activity does not show up in the statistics.
Gordon (2016) takes a rather more pessimistic view. In his book The Rise and Fall of American Growth, he argues that today’s innovations are simply having less of an impact on productivity than the innovations of the past. The change from horse and cart to an automobile is much greater than the change from an older model of smartphone to a newer model. In his view, the current rate of TFP growth is the new normal.
A final possibility is that we are on the verge of some huge technological breakthroughs that will soon show up in the productivity figures. McAfee and Brynjolfsson (2014) outline such a scenario in their book The Second Machine Age. Innovations in the field of nanotechnology and driverless cars for instance could take TFP growth to a new and higher level in the coming years.
A Fourth Industrial Revolution?
Such an acceleration of technological progress could usher in a fourth industrial revolution (WEF, 2016). We have included an example of such a development in our model for TFP growth.
The Netherlands largely benefits from innovations that happen abroad. If for instance there is a breakthrough with respect to driverless cars in the US, this leads to a strong increase in what is called the catch-up potential of the Dutch economy in our model, which boosts Dutch TFP growth.
In this case, we assume that Dutch business will then structurally increase its investment in R&D capital in order to make use of this new technology and thus contribute to higher TFP growth for the Netherlands. The government also has a part to play. We assume that the government will structurally increase its investment in human capital and entrepreneurship to benefit from the developments of the Fourth Industrial Revolution. This will subsequently also benefit productivity growth.
The cumulative effect of these developments in our model is that TFP growth will amount to 1.6 per cent in the 2018-2023 period (figure 4). To put this figure in context: TFP growth would in this case be higher than in the 1980-2008 period, when it amounted to 1.4 per cent and when developments such as the IT revolution in the first years of the 21st century occurred.
Technological progress is more than ever before a determining factor for economic growth
It is abundantly clear that the opportunities for higher potential growth lie in the field of technology. The labour supply is stagnating and the figures do not suggest that capital deepening will happen at a faster pace. Without additional efforts by the government and by businesses, the outlook for TFP growth is not yet that exciting. At the same time, a technological shock driven by disruptive innovations could change everything. We have seen this happen more than once in the past: technology can contribute more to economic growth. But this requires governments and businesses to devote more time and money to innovation and entrepreneurship. And a scenario involving higher TFP growth will never be restricted to the Netherlands alone. Other countries using much of the same technology will also have to generate higher, technology-driven growth.
We have created three scenarios for the period to the end of 2023 in order to consider these effects in more detail. In two scenarios we have used the base assumption for potential growth as described above. In one scenario, the growth of labour productivity is a full percentage point higher: the Fourth Industrial Revolution.
 For each five-year age group, we estimated the structural degrees of participation on the basis of historical data and an HP filter. We then multiplied these by forecasts for the population size of each age group. For the age group between 65 and retirement age, we assumed a degree of participation that is between the degree of participation of the 55-65 age group and the 65-75 age group.
 We have not included other recent policy measures that could increase participation, such as the Participation Act, in our analysis. For a detailed analysis of the participation effects of recent labour market policy, see the publication Kansrijk Arbeidsmarktbeleid (Dutch) by the CPB, the Netherlands Bureau for Economic Policy Analysis.
 Using an ARIMA model, we estimated the pre-2008 trend of capital deepening and projected this with the most recent actual figures (in 2015).
 The model includes independent variables such as private R&D capital, entrepreneurship and human capital that account for TFP growth. See Erken et al. 2016. We extrapolated these independent variables in order to predict future TFP growth. For instance, we assume growth of private R&D capital of 1.5 per cent for the period after 2016 and we expect the degree of entrepreneurship and the level of human capital to develop on trend.
Bhageloe, R. (2016). Recovery at last? Rabobank Special.
CPB, the Netherlands Bureau for Economic Policy Analysis (2016). Kansrijk arbeidsmarktbeleid. The Hague CPB
CPB, The Netherlands Bureau for Economic Policy Analysis (2011). Werkloosheid en de Grote Recessie. The Hague CPB
Gordon, R. (2016). The Rise and Fall of American Growth: the U.S. Standard of Living since the Civil War. Princeton University Press. New Jersey.
McAfee and Brynjolfsson (2014), The Second Machine Age: Work, Progress and Prosperity in a Time of Brilliant Technologies, New York: Norton & Company. New York:
Mokyr, J. (2014). Secular Stagnation? Not in your Life, in Secular Stagnation: Facts, Causes and Cures, CEPR Press: Washington.
Solow, R. (1957), Technical Change and the Aggregate Production Function. Review of Economics and Statistics 39 (3): 312-320
World Economic Forum (2016). The Fourth Industrial Revolution: what it means, how to respond. Blog at www.weforum.org.