Comecei a parte I com a imagem dos "caretos" da região de Trás-os-Montes como símbolo de um tempo em que cada região tinha a sua roupa, a sua gastronomia, os seus dialectos e sotaques, sem a pressão de uma uniformização centralizadora.
No final da semana passada, numa caminhada matinal junto à foz do rio Douro
Antes de mais recordemos esta evolução na economia, do século XX para o século XXI, representada pela alteração da paisagem competitiva de um pico único para múltiplos picos:
"Competition on Smooth Landscapes [Moi ici: O mundo do século XX, o mundo do mercado homogéneo]First, we consider performance in a smooth consumer landscape. ... in this type of landscape all consumers have similar preferences so there is only one large niche with a single global peak. In Figure 2, Panel (A) shows how average firm performance develops in the smooth landscape over time for different levels of competition. The learning process is clearly visible: firms start from their randomly assigned positions and gradually increase their performance as they explore the landscape and locate the global peak. After approximately 40 periods average performance has stabilized. The influence of competition is also clear: for a rising number of firms (F) the average performance stabilizes at a lower level. Thus, competition is detrimental to performance, which is exactly what we would expect in a real world setting. [Moi ici: A imagem clássica da competição económica. Quanto mais concorrentes, pior o desempenho]...firms‟ performance increases over time because they gradually approach the global peak: the distance initially drops sharply and then stabilizes after about 40 periods. This is what we would expect in an NK model with low ruggedness . However, in standard NK models without competition all firms would locate exactly at the global peak, whereas in Panel (B) an increasing number of competitors causes firms to locate at an increasing distance from the global peak. This is due to the difference between competition in horizontal and vertical differentiation ... if a firm has already located at the peak it becomes less attractive for all others. A corollary of this result is that with increasing competition the product designs offered in the market display greater heterogeneity: on average firms produce product designs that are similar but not identical to the modal consumer preference....This means that in competitive markets with homogeneous consumer preferences firms never settle down but instead continue to adjust their product designs. Because firms only move when they can increase their (expected) performance and we know from Panel (A) that average performance has stabilized, that must mean that they are engaged in a constant process of stealing each other‟s customers.This raises yet another question: which firms are moving? Does one dominant firm settle down on the peak while the others move around collecting the scraps, or does competition continue to threaten all firms? Panel (D) in Figure 2 shows that the latter is the case. It shows the proportion of market leaders, i.e. the firms with the highest market share, which are overtaken („dethroned‟) in each period. With an increasing number of competitors it becomes increasingly likely that the most successful firm will be dethroned: defending a leading market share becomes increasingly difficult the more competitors there are in the market.Taken together these results suggest that on smooth landscapes, increasing competition causes markets to become increasingly and persistently volatile. Firms do not settle down with stable product designs but rather dance around the peak, continuously jostling for the best positions and being thwarted by their competitors. In terms of products the result is a continuous stream of new but similar product designs which become more and more diverse as competition increases. The cutthroat competition of stealing market shares in these markets is detrimental to firm performance, not only on average, but even for the most successful firms who are in constant danger of losing their leading position."
Agora o outro modelo económico, aquele a que chamo de Mongo, ou o Estranhistão, o modelo em que o século XXI se está a transformar:
"Competition on Rugged LandscapesWe have established that in „smooth‟ markets, competition leads to persistently volatile processes of adaptation. We now investigate how these dynamics are influenced by different distributions of consumer preferences. Figure 3 shows average results for different levels of ruggedness along the horizontal axis (from a single niche to many niches) and competition for the different lines (one, two, four and eight firms). The results are taken from the final period in the simulation ( ). Note that as illustrated by Figure 2 this is more than enough time for the results to settle into a pattern, whether static or volatile.Panel (A) in Figure 3 shows how average performance changes with changing ruggedness and competition. For landscapes with few peaks (low ) increasing competition is detrimental to performance. This is the same result we saw in Figure 2. Here however, we see that as the landscape becomes increasingly rugged, the detrimental effect of competition on performance decreases: [Moi ici: Aquilo a que há anos designo aqui por "Live and let live"] evidently, if there are several niches it matters less if there are lots of rivals....We already know that on smooth landscapes more competition causes firms to locate further away from the nearest peak. In Panel (B) we see that as the landscape becomes more rugged, firms locate closer to the nearest peak, regardless of the number of competitors in the market. This result suggests that firms may be dispersing to serve different niches. However, this result must be interpreted with caution because in more rugged landscapes there are also simply more peaks around. Note that firms that are alone in the landscape locate slightly further away from the nearest peak as K increases from zero....what is the dynamic driving competition? One possibility is that there is constant movement both on smooth and rugged landscapes alike, with firms jostling each other off the peaks. In that case the differences in results for high levels of ruggedness may be due to the fact that the alternatives are more attractive: displaced firms can find other attractive niches to serve. Another possibility is that there is simply less movement on rugged landscapes because firms disperse and „settle down‟ to stable situations where each serves a local niche.Panel (C) suggests that the latter explanation is more likely. For markets with few consumer niches competition has a large influence on volatility: the more firms in the market, the more movement we observe. As ruggedness increases, the average number of moves per firm and period decreases, regardless of the number of competitors. In very rugged landscapes (K=9) it makes hardly any difference whether there are two or eight firms in the market: firms have reached an essentially stable distribution.Panel (D) corroborates this finding. On smooth landscapes the probability that the market leader will be dethroned depends heavily on the number of competitors. Thus, if there is a single large consumer niche then it will be difficult for any one firm to defend a lead in the market. As the number of niches increases, the number of competitors matters less and less: in the extreme case (K=9) the market leader has more than a 95% chance of defending its position even if there are eight competitors in the market. In these cases firms have dispersed to serve individual niches (local peaks in the consumer landscape) and are unlikely to move. That means firms which have located favorable niches with high performance (relative to their competitors) are unlikely to be overtaken.To summarize, the distribution of consumer preferences matters for the dynamics of competitive positioning: On smooth landscapes we observe firms constantly jostling for competitive advantage around a few peaks. Competition is detrimental to performance and even successful firms are constantly in danger having their customers stolen by rivals. As the landscape becomes increasingly rugged, firms disperse more and more. Instead of clustering at some distance around a single peak they spread out to serve individual consumer niches. This has the additional effect that in very rugged markets movement drops to a minimum, and it is very unlikely that successful firms will be overtaken. Note that this result does not happen suddenly when the number of peaks becomes greater than the number of firms (at approximately ), but occurs gradually as ruggedness increases."
Sem comentários:
Enviar um comentário