Alors que la lutte contre la corruption est généralement considérée comme une responsabilité de la Justice ou des agences anticorruptions, des recherches montrent que les institutions supérieures de contrôle des finances publiques (ISC) peuvent aussi jouer un rôle essentiel. Cependant, ce rôle n'est pas toujours clairement défini et pourrait diverger selon les contextes. L'objectif de cet article est de mieux cerner le rôle des ISC dans la lutte contre la corruption du point de vue des acteurs directement concernés, soit les membres des ISC. Treize entretiens, réalisés avec des vérificateurs au Québec et des magistrats en France et au Sénégal, démontrent une contribution effective de ces institutions dans le combat contre la corruption, bien qu'elles n'en aient pas le mandat explicite.

Supreme audit institutions (SAIs) are touted as an integral component to anticorruption efforts in developing nations. SAIs review governmental budgets and report fiscal discrepancies in publicly available audit reports. These documents contain valuable information on budgetary discrepancies, missing resources, or may even report fraud and corruption. Existing research on anticorruption efforts relies on information published by national-level SAIs while mostly ignoring audits from subnational SAIs because their information is not published in accessible formats. I collect publicly available audit reports published by a subnational SAI in Mexico, the Auditoria Superior del Estado de Sinaloa, and build a pipeline for extracting the monetary value of discrepancies detected in municipal budgets. I systematically convert scanned documents into machine-readable text using optical character recognition, and I then train a classification model to identify paragraphs with relevant information. From the relevant paragraphs, I extract the monetary values of budgetary discrepancies by developing a named entity recognizer that automates the identification of this information. In this paper, I explain the steps for building the pipeline and detail the procedures for replicating it in different contexts. The resulting dataset contains the official amounts of discrepancies in municipal budgets for the state of Sinaloa. This information is useful to anticorruption policymakers because it quantifies discrepancies in municipal spending potentially motivating reforms that mitigate misappropriation. Although I focus on a single state in Mexico, this method can be extended to any context where audit reports are publicly available.

The United States has traditionally lagged behind Europe in the adoption of voluntary or legislated standards for the care and treatment of animals on farms. US federal legislation of farm animal practices is minimal, confined to aspects of livestock transport and slaughter. Although some of the livestock and poultry producer (commodity) groups wrote guidelines, codes of practice, or statements regarding the humane treatment of animals in the 1980s, these were usually very general statements of current industry practice, developed with little consultation with independent experts and involving no mechanism for encouraging or ensuring compliance by producers. However, this has changed dramatically in the last few years, with an increasing trend among US retailers to require their suppliers to adopt minimum animal welfare standards. The major chain restaurants and supermarkets are working through their trade organisations, the National Council of Chain Restaurants and the Food Marketing Institute respectively, and with the commodity groups, to develop a uniform set of standards and a national auditing program. Standards and auditing programs have already been approved for dairy cattle, laying hens and meat chickens, and for slaughter, including ritual slaughter (kosher and halal). The process of setting auditable standards is complicated by the lack of legislative underpinning, the scope of the auditing that will be required because of US farm sizes and the large distances between farms, and the varying levels of expertise of potential auditors. For these reasons, ‘engineering-based’ auditing criteria that are relatively easy to measure and to standardise are more common. There are both strengths and potential weaknesses of retail-driven rather than legislatively driven animal welfare standards. Regardless, the recent changes in the US possibly pave the way for increasing dialogue between Europe and the US on farm animal welfare issues.

A welfare audit that utilises numerically scored, animal-based outcome measures has been used successfully by McDonald's and other restaurant companies for over ten years. In 2010, audit data from two restaurant companies indicated that all 30 of their North American plants rendered 95% or more of the cattle insensible with a single shot from a captive-bolt gun. Eight pork plants that used electrical stunning placed the tongs correctly on 99% or more of the pigs. All animals were insensible prior to hoisting. In 32 beef plants, the percentage of cattle vocalising in the stunning area was 5% or less. In 94% of the beef plants and 86% of the pork plants, none of the animals fell during handling. The worst falling score was 2% in two of the plants. High standards were attained by making simple changes. To improve welfare, plant managers did the following: improved stunner maintenance; installed non-slip floors in stun boxes and unloading ramps; and trained employees. To reduce balking and improve animal movement, the following modifications were made: illumination of dark race entrances; moving of lamps to eliminate reflections; reducing equipment noise; stopping employee yelling; installation of solid sides on races or shields to prevent animals from seeing activity outside the facility; and the elimination of air blowing in the faces of approaching animals. Employees were trained to use behavioural principles of animal handling such as the point of balance and the flight zone. The five numerically scored outcome measures in this audit are critical control points that can detect a variety of problems. They are: i) the percentage of animals stunned effectively with a single application of the stunner; ii) the percentage of animals falling during handling must be 1% or less to pass; iii) the percentage of pigs or cattle vocalising (moo, bellow, squeal) in the stun box or while entering into the stun box must be 5% or less to pass (vocalisation scoring is not used for sheep); iv) the percentage of animals moved with an electric goad; and v) the percentage of animals rendered insensible before hoisting must be 100% to pass an audit. An animal is scored as either silent or as a vocaliser and whether stunned correctly with a single application or not stunned correctly. The audit also contains a list of banned practices that will result in an automatic failure. To maintain improvements in handling, 23 plants have installed video cameras that are monitored by auditors viewing the footage over the internet. These external auditors perform numerical scoring at random times throughout the day. Video auditing over the internet is an important new tool for improving welfare.

Electoral accountability is widely considered to be an essential component for maintaining the quality of a polity’s institutions. Nevertheless, a growing body of research has found weak or limited support for the notion that voters punish political corruption, a central but partial aspect of institutional quality. In order to capture the full range of institutional dysfunction an electorate should be incentivised to punish, I further the concept of institutional performance voting, that is, voting on institutional quality as a whole. Using a novel data set on performance audit reports in Swedish municipalities between 2003 and 2014, I find that audit critique is associated with a statistically significant but substantively moderate electoral loss of about a percentage point for mayoral parties, while simultaneously associated with a 14 percentage point decrease in their probability of reelection.

Certification initiatives are product-focused, rely on standards and use sustainabilitymetrics to inform end-users on the provenance of commodities. In the metals sector, thephenomenon of formal certification programs has recently gained traction. Four initiativesare reviewed to illustrate the status and prospects of metal certification. The prime caseis the Conflict Free Smelter Program operated by the global electronics industry. Thisscheme has developed and applied standards on mineral chain-of-custody, including use ofthird-parties to audit smelters and refineries all over the world. Additional programsdiscussed are the Green Lead Project, Fair Trade and Fair Mined gold, and the ResponsibleJewellery Council. Collectively these initiatives address a variety of sustainabilitycriteria, including social, economic and environmental dimensions, but focus only onprecious and specialty metals (Au, platinum group, Pb, Sn, Ta and W). Metalscertifications programs are building capacity and infrastructure compared to matureprograms in agriculture and other commodity sectors. Opportunities and issues for growthof metals certification are considered.

Legislators have been using audit and financial crises as excuses to introduce additional regulation into an industry that is already over-regulated. This practice is questioned here because of the ability of the free market to punish audit failures, the dynamism shown by participants in relation to the voluntary adoption of new self-regulatory policies and the high costs and dubious effectiveness of the new regulations adopted. A more prudent approach that would give the market time to discover the efficient mix of services, quality safeguards and firm structures is advised. Current regulatory tendencies, the main element of which is mandatory auditing, risk condemning financial auditing to triviality and increasingly ineffective regulation.

The Sarbanes-Oxley Act and the Securities Exchange Commission move too quickly when they prod the Financial Accounting Standards Board, the standard setter for US GAAP, to move immediately to a principles-based system. Priorities respecting reform of corporate reporting in the US need to be ordered more carefully. Incentive problems impairing audit performance should be solved first through institutional reform insulating the audit from the negative impact of rent-seeking and solving adverse selection problems otherwise affecting audit practice. So long as auditor independence and management incentives respecting accounting treatments remain suspect, the US reporting system holds out no actor plausibly positioned to take responsibility for the delicate law-to-fact applications that are the hallmarks of principles-based systems. Principles, taken alone, do little to constrain rent-seeking behaviour. In a world of captured regulators, they invite applications that suit the regulated actor's interests. Rules, with all their flaws, better constrain managers and compromised auditors. Broadbrush reformulations of rules-based GAAP should follow only when institutional reforms have succeeded.