The Long Norwegian Boom: Dutch Disease After All?.

• Author: Mork, Knut Anton

1. INTRODUCTION

   In the debate about the macroeconomic effects of natural-resource discoveries, the literature appears to be unanimous in holding up Norway as a beacon of success. It has neither been plagued by the resource curse, i. e. weak or negative GDP growth due to rent seeking with weak institutions, nor by the Dutch disease, i.e. manufacturing decline and productivity weakness with subsequent loss of competitiveness. That, at least, was the verdict by the end of the 20 century. This paper asks the question of whether this success continued through the post-2000 oil and gas investment boom. The answer is in the negative. By looking at post-2000 data and moving the focus from quantities to relative prices and wages, the paper finds symptoms of the Dutch disease. Its basic message is that the investment wave produced a boom that lifted product prices and wages to levels that will be unsustainable once the resource boom is over. Furthermore, the plan to have the government collect the entire resource rent failed as about half of the rent in this period leaked out to the private sector.

   In terms of basic economics, the discovery of a valuable natural resource should be good news for the owner as well as the economy of which the owner is part, evidence of which was found by e.g. Sala-i-Martin et al (2004) and Brunnschweiler and Bulte (2008). However, a large body of literature presents evidence to the contrary. The resource curse, as documented by e.g. Ross (1999, 2015) and Frankel (2012), is mostly associated with rent seeking, corruption, and armed conflicts (Andvig and Moene, 1990; Collier and Hoeffler, 2004; Fearon, 2005; and Robinson, Torvik, and Verdier, 2006). The explanations center mainly around institutions, which may have been too weak to prevent rent seeking (Mehlum, Moene, and Torvik, 2006), or whose strength may have been hampered by resource discoveries (Sachs and Warner, 1995, 2001).

   Whereas the resource curse is typically a problem of developing economies, Dutch disease is an issue for advanced economies rich in natural resources, especially non-renewable ones because of the temporary nature of their harvesting. As is well known, the name is derived from the experience following the discovery and development of natural-gas resources in Groningen in the Netherlands. Briefly put, the problem consists of a loss of competitiveness in export-oriented manufacturing, caused by real appreciation as natural-resource revenues boost the demand for services and crowd out manufacturing, which in turn is hurt by weakened learning by doing. These issues have been analyzed thoroughly by Corden and Neary (1982), Corden (1984), van Wijnbergen (1984), and Krugman (1987). (1)

   Dutch disease seems to be preventable, however; and Norway has been held up as an example of success. That has been the virtually unanimous verdict of Norwegian (e.g. Bj0rnland, 1998; Larsen, 2005 and 2006; and Holden, 2013) as well as non-Norwegian authors (Gylfason, 2002; Thurber, Hults, and Heller, 2011; and Ramfrez-Cendrero and Wirth, 2016). Larsen (op. cit.) found that the discovery of oil in Norway accelerated GDP growth, allowing the country to overtake its neighbors Sweden and Denmark during the 1980s. Bj0rnland (op. cit.) looked at the effects of oil discovery on manufacturing employment in Norway and the United Kingdom, finding some evidence of Dutch disease in the United Kingdom, but none in Norway.

   The Norwegian success has partly been ascribed to fortuitous coincidences, such as the high state of development of the Norwegian economy at the time of the oil discovery (e.g. Ramirez-Cenero and Wirth, op. cit), and the easy conversion of domestic skills from shipping and fisheries to offshore exploration and production (Nerheim, 1992a, b, Holden, op. cit.). As a late entrant, Norway was furthermore able to learn from other countries' experience (Vislie, 2017).

   However, the main emphasis has been on the policy regime that the government established in response to the resource discovery. A legal framework of government ownership and control was developed already in the 1960s (Vislie, op. cit.; Gylfason, op. cit.) and formed the basis for the development of key institutions (Larsen, 2006; Thurber et al., op. cit.) that allowed the government to strike a balance between multinationals' know-how and its transfer to domestic agents. A tax system was established that was sufficiently market oriented to maintain oil company interest and at the same time designed to allow the government to appropriate the entire resource rent. As these flows grew large enough to matter, they were shunted into a sovereign wealth fund, the Government Pension Fund Global (GPFG). The final step, taken in 2001, was the introduction of the Fiscal Rule limiting the government's spending from the fund to the expected real return on the fund's assets (Gylfason, op. cit; Ramirez-Cenero and Wirth, op. cit; Larsen, 2005; and Holden, op. cit.). (2) Some authors, like Larsen (2006) and Gylfason (op. cit.) add more general policies on education and research, as well as the tripartite cooperation on incomes policy, which they claim prevented wages from rising out of control. Dyrstad (2016) emphasized this point especially.

   The acceptance of this narrative seems to have been universal, at least through the end of the 20" (1) century. However, the big boom of oil and gas activity in Norway did not take place then, but during the subsequent two decades. This paper exploits the fact that we now have complete data for this period. Stimulated by the combination of improved technology and an unprecedented strengthening of the world oil market following the powerful emergence of China, oil and gas investment activity during this period more than doubled from an already high level between 1999 and 2013 and remained high through 2019, as shown in Figure 1. At the peak it made up 43 percent of all fixed-investment activity in the country and 9 percent of mainland GDP. This paper asks what kind of mark this wave left on the overall economy, in particular, whether the boom it created set wages and product prices off on an unsustainable path from which the economy may have trouble returning once the boom is over. It also asks whether the government succeeded in its plan to avoid a boomtown spending spree by collecting all the rent as taxes and accumulate it in the GPFG. The answer is that it did not.

   Two papers that have looked at the post-2000 data conclude that the success against Dutch disease remains in place. Bj0rnland and Thorsrud (2016) and Bj0rnland, Thorsrud, and Torvik (2019) find that productivity in the traded-goods sector has not only continued to grow but has slightly outpaced that of neighboring Sweden since 1970. Their explanation is learning by doing as technologically advanced oil companies place orders with domestic manufacturing firms.

   This paper does not contradict their finding. Although noting that the productivity advantage over Sweden disappears when the scope is widened to the entire non-oil economy, it shifts the focus from production and productivity to product prices and wages. It finds that oil companies' demands have bid up product prices and wages much more than in neighboring Sweden after 2000 and that these increases have greatly outpaced consumer prices, so that living standards have been lifted significantly. That is good news. The not so good news is that this is exactly a case of the real appreciation that is uniformly identified as typical of the Dutch disease. In particular, downward wage stickiness (Holden and Wulfsberg, 2009) is likely to hamper the adjustment back to normal levels once the boom is over.

   Moreover, this paper finds that the effect of oil-company input demands has been exacerbated by raised household spending resulting from the increased incomes of households employed in the non-oil economy. This effect constitutes a major leakage from the government's plan to soak up the entire resource rent in the form of taxes and dole it out only gradually and sparingly in the form of contributions to the national budget. The leak comes from the oil companies' deductible costs when the prices of goods and services that oil companies demand are driven up by the positive shock to these demands. For the period 2000-2019, I estimate that roughly one half of the resource rent was diverted to the private sector via this mechanism. That is unfortunate from the microeconomic point of view of a government that had intended this flow to be collected and controlled by the public sector and spent on public rather than private goods. Macroeconomically, it would not be a problem if households saved this temporary windfall in a manner similar to the government's plans for the GPFG. However, the findings of Halonen and Mohn (2018) suggest otherwise.

   Taken together, I interpret these findings as evidence that the Norwegian government's efforts to steer clear of the Dutch disease, which had seemed to work so well through the 1990s, fell short of its targets during the first two decades of the 21 (s) ' century. This does not mean that the economy did badly during those years. To the contrary, the long Norwegian boom was spectacular, and the wealth generated was very real. However, about half of it ended up in hands it was not supposed to, politically speaking. Furthermore, the effects of the boom have made the transition to a post-petroleum economy more difficult.

   The timing of the boom is worth noting. By the turn of the century, oil and gas had been lifted for almost 30 years and had, by the mid-1990s, reached about the same level as it has currently (cf. Figure 1). Although activity, compared to neighbors, had been good in the preceding three decades, as noted by Larsen (op. cit.), the investment boom of the two most recent decades has been unprecedented. At least three forces were at work. First, the emergence of China raised oil prices and...