Produtividade? Há muitas! (parte I)

Já há algum tempo que o Bruno Fonseca não aparece pelo blogue a comentar. Foi dele que me lembrei ao ler um working paper publicado já este ano pelo Banco de Portugal, "Productivity and organization in portuguese firms".
.
O Bruno e eu fazemos alguma faísca quanto ao tema do crescimento das empresas. As empresas têm (?) de crescer mas tenho algum medo do crescimento. Quando olhamos para o sector exportador português vejo dois tipos de empresas, as empresas como a petrolífera ou as de pasta de papel que produzem commodities onde a escala é fundamental para competir e, as empresas-tipo do calçado, do têxtil e vestuário, do mobiliário, da metalomecânica, da ... empresas que não competem pelo preço e que nos últimos dez anos emagreceram bastante porque deixaram de competir pelo preço e passaram a basear-se em outras vantagens competitivas. Quando vejo uma PME a pensar em crescer muito e muito depressa penso logo que serão tentadas a apostar no preço, o que não me parece vantagem relevante para uma PME portuguesa.
.
O working paper acima citado vem dar-me razão:

"We show that as a result of an exogenous demand or productivity shock that makes the firm reorganize and add a management layer, quantity based productivity increases by about 4%, while  revenue-based productivity drops by more than 4%. Such a reorganization makes the firm more productive, but also increases the quantity produced to an extent that lowers the price charged by the firm and, as a result, its revenue-based productivity."

Na linguagem dos autores, a PME-tipo quando aposta no crescimento pode aumentar a sua quantity-based productivity mas baixa a sua revenue-based productivity. Para quem aposta em subir na escala do valor para poder aumentar preços sem perder clientes:

"The first issue is whether we want to measure quantity or revenue-based productivity. The distinction is crucial since the first measures how effective is a firm in transforming inputs and factors into output, while the other also measure any price variation, perhaps related to markups, that result from market power. The ability of firms to determine prices due to some level of market power is a reality that is hard to abstract from."

Um exemplo:

"we can use the example of a single-product firm producing aluminium cookware (anonymous given confidentiality requirements). It increased its workforce over time and, in particular, by 27 percent between 1996 and 1998. [Moi ici: Estamos a falar de um período em que a concorrência da China ainda não se fazia sentir e, se calhar, o sector ainda vivia a bonança que era ser a "china da Europa" quando ainda não havia China] In the same period exports increased by 170%, and went from representing 10% of the firms sales in 1996 to 16% in 1998. Between 1997 and 1998 the firm reorganized and added a layer of management.

Our firm had a layer of workers and a layer of managers until 1997 and it added a new layer of management in 1998 (so it went from 1 layer to 2 layers of management). Figure 3 plots its quantity-based and revenue-based productivity around the reorganization (we plot 3 alternative measures of revenue-based productivity). The pattern in the figure is typical in our data. The year in which the firm reorganizes its quantity-based productivity clearly jumps up and its revenue-based productivity declines. In contrast, it is hard to see any significant pattern in the changes in these measures of productivity for the year before or the year after adding the extra layer.

Figure 4 shows the corresponding levels of output, prices and revenue for the same firm and time period. The graph shows how, in fact, the increase in quantity-based productivity is accompanied by an increase in quantity, a fairly large decrease in price, and a small increase in revenue. These changes align exactly with our story in which the increase in quantity-based productivity generated by the reorganization (that adds a layer of management) leads to an increase in quantity, a lower marginal cost that leads to a decline in price, and a correspondingly muted increase in revenue and decline in revenue-based productivity."

Pode-se aumentar a produtividade reduzindo o custo unitário (mexendo no denominador), e/ou aumentando o preço (mexendo no numerador).

Continua.