un essai sur la crise (en anglais)

qui m'a été inspiré par un article récent de Thomas Philippon. Le texte ci-dessous devrait passer ces jours-ci sur vox-eu. Comments welcome.

A Kyoto-like permit trading system for modern finance
 
As with all human activities, modern finance has expanded to such an extent that, it generates both immensely large efficiency gains and also its share of negative externalities, namely systemic risks. The often-cited paradox is that modern finance created tools that are almost perfectly capable of insuring against idiosyncratic risks, even though some of the instruments involved, such as tritrization, rely directly or indirectly on an aggregate component, such as average real estate price, that is by definition not insurable. When the market price falls, instruments holders want to sell them off massively, bankers stop lending since their loans no longer have any trade value and hence the market slows down even more. Over and above the amplification process, the externality referred to here is an informational one, with those issuing securities generating some aggregate risk without internalizing them.

A substantial number of the academic community members are now calling for increased regulation of finance. In the past these calls were often latent but sometimes self-repressed, given the fascination for the efficiency and success of finance. Although these calls now occur on a general basis there is still no consensus with respect to the form and effectiveness that regulations should have.

Some argue in favor of ex ante authorization before issuing new financial instruments. In a way this parallels the release of new drugs, (which must first be checked by authorities to verify their toxicity), even though in the present case, toxicity depends on future aggregate events and cannot be evaluated ex ante.

Others argue rightly that taxing those financial activities generating the highest amount of external risk (establishment bankruptcies leading to the bank-run phenomena such as selling toxic assets to third parties in a world of asymmetric information) would make these activities less attractive to the financial sector and thus reduce the size of the negative externalities. In technical terms this is known as the Pigouvian tax,  named after the British economist Arthur Cecil Pigou (1877-1959). Originally this idea was applied to the analysis of pollution wherein by taxing plant products that generate negative external effects within their environment, one reduces total production, which in turn reduces global discontent regarding the production from those residing in the areas surrounding the plant.

Nobel Prize winner (1991) Ronald Coase’s main contribution was to find an alternative to the Pigouvian tax. Even though this tax was considered a good theoretical solution it required too much information on production costs and the desutility for residents. On the other hand Coase’s idea was to properly define property rights. In the specific example cited, a polluting permit would be defined and either residents or firms would be endowed with rights that could in turn lead to efficient bargaining about the final outcome.

The spirit of this idea was to let the market solve its own inefficiency problems by creating new markets, and this in turn inspired the logic for global permit trading systems, which were successfully applied to the SO2 regulations in the US (Clean Air Act in 1990) and also led to a part of the Kyoto agreement.

There is some consensus on the operational superiority of permit trading systems : the regulator can choose total quantities authorized and let prices be determined by the market, as compared to taxation schemes affecting prices, even if the former are not exempt from criticisms. In modern finance however this could very well be a track worth exploring.

The basic idea would be for each financial regulator in each macro area (US, Europe) to decide on the optimal level of trading activity within each sector (real estate securities, exotic options, and so on). Each total would be calculated according to the global risk generated by the sector and then the total amount segmented into small permits and sold to financial establishments, with rebates being possibly based on transparency and, as a further future incentive, on the amount of risk they involved in generating the 2008-09 crisis, i.e., calculated as the amount required from tax payers. Those financial establishments willing to increase their share of risky activities (here we refer to global risk, not idiosyncratic risks) because they have superior knowledge or better technologies than their competitors  could then buy more trading permits on the regional market they wish to serve. The product of the tax would be accrued in the budgets of government bodies involved and partly or fully refund recent spending (specific subsidies to banks, stimulus package).

This specific solution actually offers a triple dividend: increased incentives to financial operators and the internalizing of systemic risk externality; causing certain fiscal resources to reduce the taxation of other sectors (the traditional double dividend). Finally the economy would be stabilized through providing resources for direct government intervention. On top of that, permits could easily be issued at the macro-regional level, if this is the optimal regulation level. This would leads to additional gains in terms of further coordination across open areas; with the expected natural outcome over the next few decades being the merging of permit trading markets.

Etienne WASMER (PhD London School of Economics) is Professor of Economics, Science-Po Paris, and a research fellow at OFCE, CEPR and IZA.

Comments

[Rédigé par: Etienne]

Je poste une réaction évidemment très pertinente sur cet article (en conservant l'anonymat de la personne):

dear etienne
very quick reaction. no clue as to how this could work... we know what co2 is and know how to measure it precisely. we have little clue as to what systemic risk is, and no way of measuring marginal systemic risk correctly (we are surely trying). there are a hundred ways (through different instruments) for financial actors/institutions to achieve a given risk position. how would you define asset classes (and associated trading permit limits) to avoid regulatory arbitrage?
i am sure you have something in mind. but examples would help.

[Rédigé par: Etienne]

Et une réponse.

good point, tks a lot, I guess we did not know much more about the greenhouse effect of SO2, so fixing the amount of permits was a trial and error process, it would be the same here. et, comme pour l'environnement, un signe politique qu'on s'occupe de la question. pour les classes d'actifs, les plus concernes sont sans doute ceux qui sont non assurables (eg indexes sur l'immobilier, la valeur des devises) mais je suis sûr que les financiers en savent plus. amicalement,

Etienne

[Rédigé par: Guillaume]

tritrization -> securitization

[Rédigé par: Xav]

C'est la mégalomanie des équipes dirigeantes de grandes banques et leur prise de risque inconsidérée qui nous a mené au trou noir de l'automne 2008, pas les effets de la structuration de quelques centaines d'opérateurs de marché ("traders"). Plutôt qu'une taxe complexe sur un risque systémique improbable il faut augmenter les fonds propres des banques en comparaison de leur bilans et consolider les structures de hors-bilan des paradis fiscaux.