Bob Hancké, PEACS
4 March 2025
International Cooperation and Coordination
The green transition requires deep coordination and cooperation, but coordination is not cheap. Transaction costs (think of these as the costs involved in getting to an agreement) grow exponentially according to the number of freely coordinating (and negotiating) parties. Mathematically, the problem is represented as 2EX, where 2 is the base because the acceptable arrangement for all includes the aggregation of the solutions that are acceptable to all possible dyads, and X is the number of parties involved. In the EU, coordination would theoretically involve 2E27 (over 134 million!) dyadic agreements; if each dyad needs two days to negotiate a deal, we need over 730,000 years, ie three to four times longer than homo sapiens has existed. Such an abstract calculation may sound far-fetched but think of how difficult it was a few years ago to pass a free trade agreement with Canada, while the US-EU TIPP collapsed over disagreement among EU member states. Coordination and cooperation are expensive and not fail-proof. Conversely, EU member states can operate in blocks of like-minded countries, which would significantly reduce coordination costs and impose a much shorter timeframe; however, we would still be up against a hard limit. The solution the EU developed has been not to wait until the white smoke appears but transfer coordinating capacity to a centre with governance prerogatives.
Such an upload of competencies to a (usually federal or confederal) centre, however, create its own set of problems, which are endemic to all the attempts at deep cooperation and coordination that are necessary for a successful green transition: joint-level decision traps (Scharpf 2006), and two-level games (Putnam 1988). Joint-decision traps are situations in which all parties need to agree, and the centre can only act after agreement from all or at least a large majority of sub-units. Because such an agreement is, under most circumstances, far from given because of political and policy differences between subunits (ranging from large integrated blocs of countries like the EU, over its member states to regions within countries), such complex multi-level governance systems usually end up producing somewhat disappointing lowest-common denominator solutions. The member states that are less willing to commit to extensive green measures (for any reason, but usually because domestic actors can hold up national positions) effectively use their veto to force the others down to their desired level of coordination. The hesitation of member states often results from a two-level game dynamic, in which (actors in) sub-units hold up decisions that go against them. Because of their logic, these two-level games and joint-decision traps in cooperative arrangements accentuate the initial differences between countries, and some strategically exploit those to impose their solution – ‘my way or the highway’ – which leads to either immobilism or minimal solutions (see Hancké et al. 2025, chapter 9 for details on both concepts). Who says cooperation, in this world, says hesitation.
These complexities of coordination and cooperation matter because we do not have much time to find solutions to the climate emergency, and we expect these solutions to be quite comprehensive at the same time: all or as good as all CO2 needs to be removed from current activities ‘yesterday’ to avoid climate catastrophe tomorrow (Sharpe 2023). Again, euro-optimists may wish the difficulties of coordination away, but alas, they are there. Ambitious language in treaties cannot disguise the fact that action on the ground is slow and rare. Take the relatively unambitious example over the last five (!) years of the introduction of a deadline for 100% zero-carbon cars. To this day (early 2025), it remains unclear if 2035 will be the effective final date for fossil-fuel cars, because powerful local actors in some member states have changed tack several times. Moreover, even if we were to agree in the near future on a hard, enforceable, date around 2035 (or 2040), it would have taken the better part of ten years to get there. And the current crisis of the European car industry is not likely to help things along. Even the more ambitious UK government, which had set a phase-out date for petrol-powered cars of 2030, is considering relaxing the strict implementation of its own laws. And under pressure because of economic stagnation, Germany’s incoming government and the European Commission are significantly softening their green credentials in February 2025.
Standard Setting and Coordination
A different type of coordination problem, this time between companies, runs through the industrial aspects of the green transition. Making things, brown or green, requires two or more parties to agree on how they are made – not just in the sense of agreeing on the same process and product but also, perhaps mainly, on functional specifications, i.e. what they are supposed to do. Take the simple example of a nut and a bolt: unless the size and the thread of the bolt and of the nut are literally exactly the same, they are useless for their intended purpose. They need to be standardised, i.e. they need to follow jointly agreed or imposed technical specifications. Ditto for electricity: in fact, early European cooperation after the Second World War revolved around making electricity tradeable across borders, and for that to happen electricity producers had to agree on amperage, voltage, and other characteristics to develop a single electricity market in Europe. And the more technologically sophisticated a product or a process, the more such standards matter: they are simultaneously the language that companies in a value chain develop to coordinate their activities, and a necessary condition for the emergence of a market. If everyone develops their own standard, production and consumption are severely restricted, since proprietary standards are, by nature, exclusive and therefore exclusionary: only a select few companies can produce according to the standard. In other words, if A controls a standard, A controls access to a market. But unless that market grows into a quasi-monopoly, it also excludes potential customers and suppliers unwilling to adopt the standard.
This is the nub of the industrial coordination problem. Since everybody wants to rule the market, everyone tries to control the standards – for car batteries, for example, solar and wind energy, or software that runs energy-efficient machines and local power stations. The conundrum is that coordination is necessary to integrate value chains, but the system is geared towards competition. The effect: a standards war, in which every sizeable player tries to force others to adopt their standard, with the aim of forcing the competition to buy and adopt its standards or be gobbled up, or exit the market.
Such standards wars have been with us for a long time. Some of the most famous recent examples are Thomas Edison and the battle of the AC/DC systems (about methods to transmit electricity, not rhythmic heavy metal), the VHS/Betamax video cassette standards, QWERTY versus other keyboard layouts, or Apple v Windows. All involved two (or more) technically viable, sophisticated product standards that were battling for supremacy at the expense of the other. AC won, DC pretty much disappeared from our lives; VHS won in the living room, while Betamax remained a niche product for professional afficionados until digital media came along; nobody knows Dvorak these days (an ergonomic, fast keyboard layout that avoids learning to touch type for speed); and only in the cases of Apple and Windows have two (but not more) oligopolists alternated the lead in personal computers since the 1980s.
The problem with a standards war is not just that it produces monopolies that are not always based on quality, but primarily that it can effectively hold up investment, since no investor is willing to put their money behind one standard if the chances of betting on the wrong horse are non-negligeable. The telling example of the adoption of the GSM standard for mobile telephony shows what is at stake, and why a collective solution probably requires something more than just letting the market sort things out (in which case consumers will, supposedly, flock to the best product). In the 1980s, Europe had an advantage in basic mobile telephone technology, with strong producers in Sweden, Finland and Germany. But since the future of mobile telephony appeared to be in scale not scope – i.e. many relatively cheap phones, not a few incredibly good expensive ones – investors needed to have a sense of the potential size of the market. Even though Europe was becoming one integrated market at the time, many national telephone carriers, which were all developing their own mobile networks, persisted. In this single market with distinct jurisdictions, several standards would potentially compete, and the process would enter a vicious spiral: the market remained segregated, investment remained low, and that blocked technological development.
Before this negative cycle could start, however, the European Commission stepped in with a series of technology transfer systems and related subsidies, revolving around the single GSM standard, which went on to become the global standard for mobile telephones. A standards war avoided, the European sector could grow and come to dominate the maturing mobile phone market.
It is easy enough to imagine agreed standards producing a similar growth effect in green technologies, whereby a market for green products can develop and economies of scale come into play, with their beneficial consequences on unit price and diffusion. Yet that is only part of the story. Standards are what we call ‘sticky’: once adopted widely – a necessity for stable growth, rising investment and technological development – they are difficult to change. Many of the examples above, from alternating current to QWERTY, have been with us for many years, often several decades or even a century or more, despite the emergence of sometimes better alternatives. And even shifts, as from Apple to Windows and back, are between a few players (and often result in mutually exclusive but co-existing oligopolies). This lock-in effect is exactly why companies try to corner the market through standards: once dominant, standards rarely disappear, and the company that imposed the standard usually faces a future of increasing returns to scale, thus further entrenching the standard. But in fast-moving fields this unitary dominant standard often precludes the emergence of other, perhaps more sophisticated or better ones.
Standard setting thus faces a series of conundrums: it must balance the need for uniformity to produce beneficial economic effects with the need for forward compatibility if technologies evolve – which they are likely to do in a fast-moving market. Very often old and new standards are almost mutually incompatible (and if they are compatible, many of the new standard’s advantages are lost, since performance will converge on the lowest common denominator).
How to balance those different claims is not easy to plan, and definitely not ex ante. Leaving it to the market may simply reignite a standards war – precisely what we were trying to avoid. But any form of coordinated action and deliberation is dependent on the existence of institutions that induce parties to forego potential short-term gains and trust each other sufficiently to further a common long-term goal. The life of the GSM standard is an enlightening example. There is little doubt that for twenty-odd years after the adoption of GSM, European mobile phone producers thrived. However, while they reaped the benefits of the quasi-monopoly and its early advantages, their innovations followed the trail set by that technology. The upshot: today no one under the age of 25 remembers the Finnish handset producer Nokia, or Sweden’s Ericsson, both of which collapsed after the iPhone appeared in 2007.
The final bottleneck in the green transition is perhaps also the most important one: what is the net cost of the transition? It is in the nature of the time inconsistency problem that costs are concentrated at the start, while benefits are distant, far-reaching but often diffuse and vague – not least because currently many costs are not properly identified – and cover a wide population. While it is close to impossible to calculate the net cost (i.e. costs minus benefits) with any accuracy, we have a few tools in our box that help us think through this problem. That will be the topic of my next essays, starting in a few days.
Comments