Lessons in watchdog design from Scotland and Wales

How much autonomy does a watchdog require in order to perform its function effectively? One product of the space created for fresh thinking by devolution has been the evolution of two distinct models with which to answer this question, both of which offer greater clarity than to be found in the old British constitution model.

Famously, James Madison stated that for a liberal society to flourish auxiliary precautions were needed to cover for the risks of democracy. His solution to this dilemma is one that still dominates: namely the separation of powers between the executive, the courts and the legislature. But over 200 years later, many constitutional orders around the world have taken this basic understanding and refined it considerably. The tripartite core still provides the dominating foundation, but a range of specialist institutions have been introduced to add a layer of objective scrutiny to the system. Constitutional watchdogs such as the auditor-general, ombudsman, electoral commissions, and the standards and information commissioners are now ever present and gaining sufficient longevity to suggest permanence. For most, a role and function exists that has risen above politics.

Such arguments are widely understood, but the Westminster settlement has resisted formalising the status of watchdogs. By contrast, in newer constitutions watchdogs are written into the constitution itself whilst in parliamentary democracies such as Australia and New Zealand, solutions have been devised to reflect the reliance placed upon watchdogs. It is in this respect that devolution in the UK provides an interesting insight into the future.

The Scottish approach to understanding watchdogs is to base the responsibility for them firmly within the sphere of the legislature. The Scottish Parliamentary Commissions and Commissioners Act 2010 Act grants the Scottish Parliament full responsibility for the selection, dismissal, budget and general oversight of seven legislative watchdogs. The arguments in favour of this model reflect two core design challenges that need to be built into the establishment of watchdogs: (i) that watchdogs need to operate with autonomy in order to maximise their claim to a non-political voice; (ii) that to gain legitimacy, watchdogs need to be accountable. The legislative watchdog solution is a common answer to both challenges, as it allows the legislature to act as a buffer between the watchdog and the executive, and potentially removes altogether the influence of the executive. Simultaneously the legislative watchdog solution provides a natural forum within which the operations of the watchdog can be periodically scrutinised.

The Scottish approach is attractive but fallible. The difficulty with the legislative watchdog model is that, by itself, it conceals the way that watchdogs operate and assumes that all varieties provide much the same service. In practice, there is a close linkage between the two political branches of parliamentary democracies and to make the legislative watchdog model work counter veiling conventions are needed to limit the degree of interference of the legislature into the work of watchdogs. Such conventions are easier to secure for some watchdogs than others. Over the years conventions have been developed around financial audit, independent compliant-handling, electoral and ethical integrity, issues of general constitutional concern that legislatures around the world have a long-standing history of respecting above parochial concerns.

Whether the same strength of conventions can be built up around the work of all watchdog schemes is questionable. It is in this respect that the Welsh adoption of a tiered approach to watchdog design is defensible. The Auditor-General, the Public Services Ombudsman and the Standards Commissioner are set up under the direct auspices of the Assembly, whilst an array of other commissioners are connected to the Assembly but ultimately sponsored by the Welsh Government. At first sight this looks weaker than the Scottish model, but the work of the commissioners in Wales reflects a newer generation of watchdogs focussed on promoting an ambitious rights agenda. Equivalents exist elsewhere, including Scotland, and the need for autonomy in their operation remains strong, but the goals they pursue are arguably more likely to risk embroiling the watchdog concerned in political dispute. As a result, whether legislatures will always be as prepared to respect the output of such watchdogs to the same extent as the Auditor-General or the Ombudsman is doubtful, and to include them in the same constitutional category as the Auditor-General or the Ombudsman risks watering down the brand.

The ‘old’ British constitution struggles to comprehend the status of watchdogs, but the Scottish and Welsh models help clarify the way forward. In both models, autonomy and accountability in watchdog design are givens and the Scottish model is particularly impressive in its focus on these features. But the primary goal of watchdogs should be to bring into the political domain objective accounts of public service concern and, if necessary, the executive should be made to work through the legislature in its responses to those accounts. It is not always obvious that these goals require the watchdog concerned to be solely accountable to the legislature, particularly in single chamber legislatures which might struggle to distinguish their oversight functions from their political advocacy role.

To conclude, the different solutions adopted in Scotland and Wales offer a fascinating case study for the future and hopefully a guide for the rest of the UK to follow when it finally comes to reforming the way that the Westminster settlement deals with constitutional watchdogs.