8.4. Connecting the Pieces – A Feedback Systems View
As we have seen, the project team supplied their knowledge and opinions on the pieces of the global oil industry. Now let's see how the pieces fit together. Figure 8.15 shows the main concepts necessary to describe the industry. Take a few minutes to inspect this scatter list of words and phrases. Altogether, there are 37 phrases on the page which record shared vocabulary developed from team dialogue. Some phrases are specialised industry jargon such as intended marker price, call on OPEC, agreed quota, negotiated quota, and cartel quota bias. As a list for describing a large and complex industry it is short, but as nodes of a web it is quite daunting. Potentially, there is an enormous number of ways to connect 37 concepts. In fact, the team's knowledge of industry structure greatly reduced the raw combinatorial problem. The modelling process pinned down, with some confidence, eight feedback loops that capture (at least for team members) the essence of the industry's enduring feedback structure and the basis of its dynamic complexity.
Figure 8.15. Scatter list of phrases describing the global oil industry
8.4.1. Two Invisible Hands and More
Figure 8.16 shows six feedback loops within and between the independents' and market sectors. (By the way, you can create your own industry web by photocopying Figure 8.15 and then drawing the connections.) Our review starts with commercial supply loop B1. A production shortfall stimulates a rise in market oil price and an increase in the price to cost ratio which, through capacity approval and construction, leads to expansion of independents' capacity and production. Extra production corrects the shortfall and completes loop B1 – the first invisible hand. A new loop B2 arises from reserve depletion. The price to cost ratio is influenced by development and operating cost per barrel, which rises as undeveloped reserves fall. Reserves are depleted by capacity approval, thereby completing the new balancing loop B2 that captures the economics of finite global oil reserves as described earlier in Figure 8.5.
Figure 8.16. Feedback loops in the oil producers' model
In the 1995 version of the microworld, reserves are replenished by the adoption of Russian oil. The rate of adoption depends on the magnitude of risky Russian reserves and the time to build trust in Russia which is a scenario parameter. Time to build trust can take a value anywhere in the range 5–40 years. If the time is short (say 10 years), then Russian oil fields are commercialised quickly, giving rise to a surge of commercial development and production exploiting the favourable economics of Russian oil regions. If the time to build trust is very long (say 40 years), then Russian oil, though economically attractive, is commercialised slowly.
Now we turn to the demand side to find a second invisible hand. The production shortfall is the difference between demand for oil and the sum of independents' and OPEC production. Demand for oil falls as price rises and vice versa. But consumers take time to adapt and to some extent they are addicted to oil. These behavioural factors are captured in loops B3 and R1. Demand for oil adjusts with a time lag to consumers' intended demand. Intended demand responds to the price ratio and the effect of the global economy and environment (a link we return to shortly). As the price ratio goes up, intended demand goes down. The price ratio itself combines factual and behavioural information. It depends directly on the market oil price, thereby completing loop B3 representing consumer price response. The ratio also depends on the benchmark price, which is the price consumers are used to. This benchmark adapts with a time lag to the market oil price to form loop R1 that captures price drift in consumer behaviour. The combination of feedback loops B3 and R1 is very common in business and social systems. The structure is known as an eroding goal archetype (Senge, 1990, Appendix 2) and represents the tendency in any goal-seeking, purposive enterprise for the goal itself to adapt to current conditions. In this case, consumers facing high oil prices eventually get used to them and carry on consuming – they are hooked on oil.
The effect of the global economy and environment is a scenario parameter. It sits on the edge of the feedback structure representing pressures on demand that are not directly attributable to price. It is an enormously versatile parameter. It is a single number (confined in the range −0.1 to 0.1), yet it can portray scenarios as different as a green world or an Asian boom and bust. When the parameter is set to its neutral value of 0, then price alone drives intended demand. Market forces prevail on the demand side. When the parameter is set to a value less than 0 there is continual pressure from consumers to reduce demand, independent of market forces (though market forces are of course still at work). Such downward pressure might correspond to environmental awareness (a green mindset), the effects of an economic recession or steady increase in the efficiency of energy consuming devices from technological progress. The exact cause need not be modelled in detail. Whether driven by the global economy or the environment, the effect is to suppress demand below what it would otherwise be on the basis of price alone. Conversely, when the parameter is set to a value greater than 0, there is continual pressure on consumers to increase demand, due to boom times and general optimism about the future.
8.4.2. The Visible Hand of OPEC
Loops B4 and R2 complete Figure 8.16 and show OPEC's control of oil price exercised by the swing producer. OPEC production in the bottom right of the figure is the sum of swing producer production and opportunists' production. When the member states are in harmony they produce to quota, so in each case production depends on negotiated quota. Two circumstances can cause the swing producer to depart from negotiated quota. One is punitive action, but this is quite rare and occurs only when the swing producer's share of production is too small (see next section on 'webs of intrigue'). The most common circumstance is a short-term tactical adjustment of production to manage price – a legitimate and important role for the swing producer. In Figure 8.16, swing producer production is influenced by the price gap, which is the difference between intended marker price and market oil price. This connection to market oil price closes a balancing feedback loop B4 that passes back through production shortfall and OPEC production before reconnecting with swing producer production. This fast-acting balancing loop represents OPEC's price control and is capable of creating prolonged periods of price stability, as seen in the 1960s. When market price falls below the intended marker price, due to a temporary supply glut, the swing producer quickly cuts production below negotiated quota to bring price back in line with the target or marker. The loop acts quickly because the swing producer is willing to make capacity idle – a process that takes only a month or two. Similarly, when market price rises above the intended marker price, due to a temporary demand surge, the swing producer quickly re-activates idle capacity to increase supply and bring price down. Despite the popular bad-guy image of Saudi Arabia, the swing producer is in fact a benign and calming influence in global oil markets, boosting or curtailing production in order to maintain stable prices. As we shall see, this benign role only becomes sinister and threatening when OPEC as a whole agrees quotas that deliberately undersupply or oversupply the global oil market, or when the swing producer is provoked into punitive mode.
The intended marker price itself changes over time. It depends on the recent history of market oil price. This adaptive connection from the market price to the marker price completes a reinforcing loop R2 (OPEC price drift) that combines with loop B4 to form another eroding goal archetype.
8.4.3. Webs of Intrigue – Inside OPEC's Opulent Bargaining Rooms
Figure 8.17 shows feedback loops within OPEC's quota agreements and quota negotiations. The starting point is the call on OPEC, which is the difference between demand for oil and independents' production. OPEC member states use the call as a first-cut guide to their agreed quota. This approach makes sense because when quota equals call then OPEC is exactly filling the supply gap between global demand and independents' production. But OPEC can manipulate quota above or below the call depending on political and social motives reflected in the cartel quota bias. Quota bias is a scenario parameter that can be set to represent an OPEC supply squeeze or even political turmoil in OPEC. When the parameter is set to its neutral value of 0, then agreed quota is equal to the call. When the parameter is negative, say −0.1, then agreed quota is 10 per cent less than the call and there is a sustained supply shortage or squeeze (as happened in the 1970s). When the parameter is positive, say +0.1, then agreed quota is 10 per cent greater than the call and there is a supply glut (as happened in the late 1990s).
Figure 8.17. More feedback loops
Agreed quota is allocated between the swing producer and the opportunists through negotiation and hard bargaining. Opportunists' negotiated quota depends on agreed quota and opportunists' share of quota. Negotiated quota then drives opportunists' production and also justifies expansion (or contraction) of opportunists' capacity. Capacity is a surrogate for bargaining power. The more capacity OPEC members bring to the bargaining table, the stronger is their case for a bigger share of quota. This bargaining logic is captured in opportunists' share of quota that depends on opportunists' capacity relative to OPEC capacity as a whole and completes reinforcing loop R3 in which successful quota negotiations lead to more capacity, more bargaining power and so on. OPEC capacity is simply the sum of opportunists' capacity and swing producer's capacity, where capacity is defined (for the purposes of bargaining) as operating capacity excluding any idle capacity. Loop R4 is the mirror image bargaining process of the swing producer. Here, negotiated quota depends on OPEC agreed quota and the complement of opportunists' share of quota. As before, quota drives capacity, which affects bargaining power and share of quota.
The combination of loops R3 and R4 is a common feedback structure known as the success-to-the-successful archetype (Senge, 1990, Appendix 2). This archetype occurs repeatedly in business and social systems where rivalry is at play. A successful player reinforces its competitive position at the expense of rivals. The phenomenon is well-known in battles between competing industry standards such as Betamax versus VHS in videos, or Windows versus Macintosh in operating systems. The dominant standard attracts more followers, which increases its dominance. Similarly, in quota bargaining, the actions of the successful negotiator strengthen future bargaining position while the rival becomes ever weaker.
Opportunists are easily tempted to boost their bargaining power by covert additions to capacity. For some member states of OPEC, the temptation is huge. They are developing economies with large populations, inadequate infrastructure and endless possibilities for revenue-consuming development such as social investment programmes in education, health, housing, road building and so on. In the model covert capacity comes from opportunists' capacity bias.
Capacity bias is the last of the scenario parameters in the model. It is used to invoke scenarios such as political turmoil in OPEC and quota busting. When the parameter is set to its neutral value of 0, then opportunists adjust capacity to match negotiated quota. When the parameter is positive, say +0.1, then opportunists covertly add 10 per cent capacity more than quota alone could justify in the expectation that the extra capacity will bring them more bargaining power. This covert investment provides the opportunists with surplus capacity above quota.
If opportunists' share of quota becomes too large then the swing producer takes punitive action by greatly expanding production and abandoning quota. The result is a temporary supply surplus which drives market price down. If market price falls significantly below the intended marker price then this price gap is a clear signal for opportunists to stop using their surplus capacity and to fall back in line with quota.
The scatter list of 37 phrases in Figure 8.15 has become a network of more than 50 interconnections and eight dynamically significant feedback loops in Figures 8.16 and 8.17. This web embodies a slice of dynamic complexity from our modern industrial society with its intricate membrane of stakeholder interdependence. The best way to investigate this complexity is through simulation. Before presenting simulations, however, first imagine how the web might work by conducting a thought experiment.