1 See Florentin Blanc, “Assessing Business Inspections’ Effectiveness and Burden”, 2011, available on the Internet at <http://www.ecpr.net.eu/MyECPR/proposals/reykjavik/uploads/papers/2136.pdf> (last accessed on 16 July 2012).

2 Chang, Ha-Joon, Bad Samaritans: Rich Nations, Poor Policies, and the Threat to the Developing World (London: Bloomsbury, 2007), at Chapter 9Google Scholar.

3 Essentially 100% of resources spent on inspecting in FSU. Confidential data from regulatory agencies in OECD countries suggests that, there, at least 20% is spent on analysis and back-office work (and regulatory work includes not just inspecting, but informing).

4 Looking at percentage of establishments would be more accurate, as a single business may operate several. In surveyed countries, however, the majority of businesses (and the near-totality of SMEs) correspond to only one establishment. Since, for sampling reasons, surveys were based on business population [because business registries are based on entities, not premises), the two are assumed to be essentially equivalent here.

5 The populations surveyed are not entirely identical, due to differences in the registration of sole proprietors (and the possibility, or not, to combine their sample with legal entities’), the inclusion or not of agricultural producers, etc. Nonetheless, the general picture is comparable. See survey reports for detailed methodologies.

6 To assess targeting, it would be better to have data on the percentage inspected out of the supervised population, i.e. the establishments that the regulator effectively has competence upon, but in most cases, for the countries considered, this population can be equated with the general business population, as regulators have very broad mandates, and make full use of them.

7 Azerbaijan: 2007, Belarus: 2003 and 2008, Georgia: 2003 and 2005, Kyrgyzstan: 2008 [farms excluded], Mongolia: 2008–9 (Q4 to Q3), Tajikistan: 2003 and 2007, Ukraine: 2006 and 2010, Uzbekistan: 2001 to 2007. Reforms started in: Azerbaijan 2011 only, Belarus 2006, Georgia 2001 with acceleration in 2003, Kyrgyzstan 2005 but stalled several times because of political events, Mongolia 2003 but stalled and resumed in 2009 only, Tajikistan 2006 with slow implementation, Ukraine 2007 with similarly difficult implementation, Uzbekistan 1998 deepened from 2002.

8 Just a couple in Mongolia, where most agencies were consolidated in a “single inspectorate” – around 20–30 in Tajikistan, Kyrgyzstan, Uzbekistan – around 80 in Ukraine. This compares with around 70 in the Netherlands or Lithuania, similarly with only a few being really active and most of them being very small, or around 15 in Latvia or Slovenia.

9 This over-estimates the percentage of businesses inspected, since some may have visited twice, and this this data is not usually available. This still allows to get an order of magnitude.

10 Survey conducted on 2011 data, for the Civil Service Department of the Presidency of the Council of Ministers. To be published.

11 See the summary note on these: Inspection Council Bureau (Netherlands), “Inspection Reform Programme”, 2009, available on the Internet at <http://www.inspectieloket.nl/Images/A5-folder-VTaug09_310809_en_tcm296–264134.pdf> (last accessed on 16 July 2012).

12 This seems largely true in the EU as well.

13 Though actual costs per inspection were far higher for legal entities.

14 Not publicly available, obtained through an MP's request to the Tax Service.

15 Combined with official data on fires from the Ministry of Emergency Situations available on the Internet at <http://www.mns.gov.ua> (last accessed on 16 July 2012).

16 Of course fire incidence is an unsatisfactory risk indicator for fire safety, since the possibility of human casualties is the main focus – but this can be used as a first indication that targeting does not seem based on risk analysis, since small cafes and restaurants are also not high-risk from the “potential fatalities” perspective.

17 In some EU countries, the split between national and local regulators may partly hide the inspections pressure. Surveys conducted by the UK's LBRO (now BRDO – see at: <http://bis.gov.uk/brdo> last accessed on 16 July 2012) did not ask about inspections but showed that over the past 2 years, 50% of businesses had some contact with local regulators (often as part of inspections). Research conducted by the Dutch Government as part of its “Inspection Holiday” initiative suggests that, in many sectors, most businesses are inspected every year – but essentially by local agencies (which inspect 2 to 20 times more often than national ones, depending on sectors).

18 See for instance summary in of Morgan, Bronwen and Yeung, Karen, An Introduction to Law and Regulation (Cambridge: Cambridge University Press, 2007) at p. 200–202 CrossRefGoogle Scholar – arguments for an enforcement approach that makes upholding compliance with the law, and enforcement thereof, a fundamental value, and thus makes other approaches to “risk” subordinate to it.

19 These observations are based on the author's personal experience while supporting the development of risk-based planning of inspections in the surveyed countries.

20 An excellent example (a contrario) is through an essay that strongly criticizes the risk-focused reduction of inspections (and prosecutions) by the Health and Safety Executive in the UK (Steve Tombs and David Whyte, Regulatory Surrender. Death, Injury and the Non-Enforcement of Law (Liverpool: Institute of Employment Rights, 2010). The authors did good investigative work to obtain data on numbers of inspections, prosecutions etc. by the HSE over several years (which the HSE should be publishing, but is not) – but fail to be able to prove that this results in more deaths and injuries, as none of the available statistics support this. Their claim that statistics are unreliable is not seriously backed up. In fact, the HSE’s risk-based approach appears vindicated if one looks neutrally at the data (even though changes in the structure of the economy and in technology may have helped as well). The effectiveness of riskfocused inspections is also supported by Prof. Ragnar Löfstedt, “Reclaiming health and safety for all”, UK Department of Work and Pensions, 2011, available on the Internet at <http://www.dwp.gov.uk/docs/lofstedt-report.pdf> (last accessed on 16 July 2012).

21 See for instance Charles Vellutini, “Key Principles of Risk-Based Audits” (ch.1), “Risk-Based Audits: Assessing the Risks” (ch.2) and “Database and IT Framework for Risk Analysis” (ch.5) in Munawer Sultan Khwaja, Rajul Awasthi and Jan Loeprick (ed.), Risk-based Tax Audits. Approaches and Country Experiences (Washington D.C., World Bank Group, 2011), pp. 13–36 and 57–64 – and also Ugur Dogan, “Data Warehouse and Data-Mining Tools for Risk Management: The Case of Turkey” (ch. 7), ibid., pp.71–76.

22 A key “reality check” is to compare the risk categories thus created to relevant statistics on hazards affecting the country, when possible), otherwise absurdity can ensue. E.g. in Kyrgyzstan hairdressers were classified uniformly as “high risk” due to old Soviet-time rules (and rent-seeking considerations), even though no health statistics backed this up. (NB this example is great, you may want to put ijn in the main text)

23 Examples of difficulties in integrating legacy systems see: UK's Retail Enforcement Pilot case study author date available on the Internet at <http://www.lbro.org.uk/resources/retail-enforcement-pilot.html> (last accessed on 16 July 2012). In the Netherlands, the e-Inspection programme has made progress but it remains incomplete because of similar challenges – see on the internet <http://www.inspectieloket.nl/vernieuwing_toezicht/programma_einspecties/> (last accessed on 16 July 2012). For an example of “best practice” see province of Nova Scotia in Canada <http://www.gov.ns.ca/snsmr/access/business.asp> (last accessed on 16 July 2012) and Mike Davis, “Designing and Implementing Business Licensing Portals”, 18 May 2011, available on the Internet at <https://www.wbginvestmentclimate.org/uploads/Session%205%20Mike%20Davis%20Designing%20and%20Implementing%20Business%20Licensing%20Portal.pdf> (last accessed on 16 July 2012).

24 In this respect there are similarities between the Soviet Union and companies such as Ikea, McDonald's or Accor, which all “standardize” all aspects of their stores, restaurants and hotels. (NB one of the main differences must be that those companies are a lot more efficient)

25 For an extended discussion of these issues see the following World Bank Group publications: IFC, “Technical Regulations in Ukraine: Ensuring Economic Development and Consumer Protection”, 2008, available on the Internet at <http://www.ifc.org/ifcext/uspp.nsf/AttachmentsByTitle/TR_Survey_ENG_2008/$FILE/0625+TRSurvey_Final_ENG.pdf> (last accessed on 16 July 2012) and IFC, “Reforming Food Safety Regulation in Ukraine: Proposals for Policymakers”, 2009, available on the Internet at <http://www.ifc.org/ifcext/uspp.nsf/AttachmentsByTitle/Food_Certification_policy_note_eng/$FILE/Food_certification_ENG.pdf> (last accessed on 16 July 2012).

26 See World Bank Group publications – for instance the “Permits” section in: IFC, “Business Environment in Uzbekistan as seen by Small and Medium Enterprises”, 2005, available on the Internet at <http://www.ifc.org/ifcext/uzbeksme.nsf/AttachmentsByTitle/SME_chapters_2004_eng/$FILE/SME_chapters_2004_eng.pdf> (last accessed on 16 July 2012) and IFC, “Business Licensing Reform: A Toolkit for Development Practitioners”, 2006, available on the Internet at <https://www.wbginvestmentclimate.org/toolkits/business-licensing-toolkit/> (last accessed on 16 July 2012).

27 See Morris Kleiner, “Occupational Licensing”, 14–4 Journal of Economic Perspectives (2000), pp. 189–202, also available on the Internet at <http://web.missouri.edu/∼podgurskym/Econ_4345/syl_articles/kleiner.pdf> (last accessed on 16 July 2012) and more recently and comprehensively: Morris Kleiner, Licensing Occupations: Ensuring Quality or Restricting Competition? (Kalamazoo: Upjohn Institute of Employment Research, 2006).

28 A valuable collection of such studies can be found in the Netherlands’ inspections portal library of effectiveness studies available on the Internet at <http://www.inspectieloket.nl/vernieuwing_toezicht/toezichtmeteffect/bibliotheek/> (last accessed on 16 July 2012) – however this includes many studies that are not strictly of “inspections” but rather of “control” in general (e.g. driving rules enforcement).

29 On effectiveness of different regulatory tools and approaches, contradictions abound: OSHA rules and enforcement in the USA are either very effective or not at all – arguing the latter are Ann P. Bartel and Lacy G. Thomas, “Direct and Indirect Effects of Regulation: A New Look at OSHA's Impact”, 28(1) Journal of Law and Economics (1985), pp. 1–25 – arguing the former are (among others) Scholz, John T. and Gray, Wayne B., “OSHA Enforcement and Workplace Injuries: A Behavioral Approach to Risk Assessment”, 3 Journal of Risk and Uncertainty (1990), pp. 283–305 CrossRefGoogle Scholar. A recent study purports to settle this dispute by using longer time series and a larger sample, and suggests OSH inspections are effective (see Levine, David I., Toffel, Michael W., Johnson, Matthew S., “Randomized Government Safety Inspections Reduce Worker Injuries with No Detectable Job Loss”. 336 Science, (2012), pp. 907–911 CrossRefGoogle ScholarPubMed) but some of the inferences in the model (and choices in the sample) may impact the results, so it is unlikely to remain undisputed, and in any case does not test different approaches to inspections planning against each other, but just inspected vs. not inspected.

30 See May, Peter J. and Winter, Soren, “Regulatory Enforcement and Compliance: Examining Danish Agro-Environmental Policy”, 18(4) Journal of Policy Analysis and Management (1999), pp. 625–651 3.0.CO;2-U>CrossRefGoogle Scholar for environment; and for tax see Kirchler, Erich and Hoelzl, Erik, “Modelling taxpayers’ behaviour as a function of interaction between tax authorities and taxpayers”, in Elferrs, Henk, Verboon, Peter and Huisman, Wim (eds.), Managing and Maintaining Compliance (Leiden: Boom Legal Publishers, 2006), pp. 1–20 Google Scholar. The former show that regulators’ approach and behaviour can have different outcomes depending on the context, relations between regulators and regulated etc., while suggesting that increased frequency of inspections leads to improved compliance outcomes. The latter exposes how there are trade-offs between stricter and more intensive enforcement, and more “trust-based” regulation, each of them becoming counter-productive when taken to an extreme.

31 Statistics for fire-related deaths by population gathered by the International Association of Fire and Rescue Services available on the Internet at <http://www.ctif.org> (last accessed on 16 July 2012) for instance show Uzbekistan having a low fatality rate. Whether this is because of structural reasons (low fire risk), under-reporting or effective enforcement cannot be ascertained. On the other hand, Russia's incidence is 10 times higher than the USA’s, and both countries are more comparable (and Russia's inspections system is very similar to our surveyed countries’), which suggests low effectiveness of Russian inspections and/or legislation may be a factor.

32 For an example of study suggesting such a positive correlation see Gray, Wayne and Scholz, John, “Analyzing the equity and efficiency of OSHA enforcement”, 13 Law and Policy (1991), pp.185–214 CrossRefGoogle Scholar.

33 Post-Soviet lawyers consider that powers are given by laws to executive branch agencies have to be implemented to their fullest at all times, and thus that reforms based on a risk-based approach, with discretion not to inspect sometimes, as illegal – even though most international legal scholarship would consider this up to the executive branch to decide to what extent to make use of powers vested into it by the legislative.

34 The situation is often as problematic in EU and OECD countries. The Inspection Council of the Netherlands counted above 70 structures authorized to inspect a few years back. Lithuania has around 80 (and is actively working to reduce this number). New agencies often get created in an RRR context, without any review of existing ones.

35 Consolidation of agencies is neither a panacea, nor guaranteed to succeed, but in this paper's perspective it is relevant as a way to reduce the “automatic RRR” tendency.

36 The Baltics show the possibility of real success, but also the time it takes to reach a full transformation. In spite of EU accession, Lithuania (as per a survey conducted for the government) still saw 60% of businesses (legal entities) inspected at least once in 2011, which is one of the drivers for its active inspections reform programme.

37 Except Turkmenistan, where no such reform has taken place so far.

38 Tombs and Whyte, Regulatory Surrender, supra note 20.

39 An example recorded recently <http://www.sudouest.fr/2012/03/16/prises-de-bec-a-la-barre-660238–7.php> (last accessed on 16 July 2012). An ornithologist prosecuted for sheltering a few birds, which he had saved from injury, but were wild species (not endangered) for which he would have needed prior approval. Two typical RRRapproaches: the inspectorate (ONCFS), instead of simply reminding him about the law, forwards the case to the prosecutor. The prosecutor, instead of dropping the case, actively prosecutes it, and appeals when a lower court rejects the charges. This costs state resources and discourages positive private activity. It results from the complete lack of a “responsive” (in the sense of Ayres and Braitwhaite’s “Responsive Regulation”) approach to inspections and enforcement, and because of the prevalence of the RRR in official policy (all infractions have to be prosecuted and punished, all the time, everywhere, was the official Government position).