Does red tape not matter for firm productivity?
This page in:
A new World Bank working paper finds that the answer, counterintuitively, is 'yes'. De Rosa et al. look at a large sample of firm-level surveys completed in 2009 and find that:
...while the payment of bribes is negatively associated with the productivity of the bribing firm, time spent dealing with bureaucratic requirements per se appears to be irrelevant.
This result is quite surprising, considering that managers of firms in many countries spend large chunks of time on pointless bureaucratic procedures. Should we give up on trying to slash regulatory burdens, contra this blog post by World Bank economist Mary Hallward-Driemeier? Not quite so fast!
There is one aspect of the De Rosa paper that leaves me unconvinced. It looks only at firm-level data from a single year. (Footnote 7 confirms this point.) But firm-level data of this type suffers from 'survival bias'. We only get responses from the firms that have managed to survive the particular regulatory and economic environment they find themselves in. In poor regulatory environments, only those firms that have managed to adapt themselves to burdensome regulation will show up in the sample. But a country that slashes red tape might see a productivity boost not only in existing firms, but also in new firms that are more suited to take advantage of a better regulatory environment.
Of course, this doesn't mean that the conclusion the authors reach is necessarily wrong. But to prove the point that red tape doesn't matter for firm productivity, they would need some kind of panel data to assess the impact of changes in business regulation on firm productivity over time. In the meantime, policymakers in countries with poor regulatory environments ought to keep their focus on slashing red tape.
(Thanks to Christian von Drachenfels for the pointer.)
Excellent point on survivorship bias. incumbent participants may be connected or familiar with policy and procedure, they may also have procured "protection" from various bureaucratic hurdles thus minimizing impacts.
The cost is innovation and entrepreneurship of new participants which gets stifled. The incumbent firms effectively can act as the captives of rent seeking bureaucrats in exchange for maintaining the red tape which hinder competitive threats to their margins.
The feedback loop in the end comes at the expense of consumers and the market economy as a whole due to the lack of innovation and drive for greater efficiencies or higher service levels by incumbent firms.
Ryan, indeed an excellent point you make and you are right that suggesting to stop cutting red tape would be a misguided interpretation of de Rosa et al.'s research.
I see the point Nick and you make about the "even more productive firms", which are maybe not there due to the costs incurred by the "time tax" when trying to start a business.
Nevertheless, the results are still puzzling to a certain extent.
Just a thought that came to my mind:
The "bribe tax" takes money away from productive activities directly as the money used to bribe cannot be spent on resources or wages.
The "time tax" is an indirect cost factor in the sense that the manager cannot supervise production, nor think about ways to increase productivity while he is dealing with public officials. But maybe this absence of the manager (The BEEPS Survey specifically asks about: "Percentage of time spent by senior management with public officials in order to obtain favourable interpretation of regulations") is not that bad for productivity?
Especially in medium and larger firms production might not as much be negatively affected by the absence of the manager as in micro or small firms, if we assume that workers in larger firms are e.g. also supervised by shop floor managers, whereas in small firms there is only one supervisor - the owner/manager of the firm.
Actually it strikes me that this might all depend on the size of a firm...I am no expert at all on econometric methods but if I understand it right de Rosa et al. do check for differences in firm size using a variable for medium (20-99 employees) and large (>99) firms. I don't know why don't check for small firms, maybe data was not available.