A social constructionist perspective on the diffusion of innovative fiscal responsibility legislation.

Author:Wallis, Joe

    There is a growing cross-disciplinary literature on the 'diffusion' of policy 'innovations' (reviewed by Weyland, 2005 and Dobbin et al., 2007) that shares overlapping research concerns with the literature on policy learning (Heclo 1974), policy transfer (Dolowitz and Marsh 2000) and lesson drawing (Rose 1991). At the very least, contributors to this literature are expected to (i) define the characteristics of the reform that render it 'innovative' and (ii) select from these a 'core' subset that can be used to establish the substantive similarity of reforms implemented in other countries so that their emergence can be represented as being part of the diffusion process of the original innovation.

    That this may be less straightforward than it appears can be illustrated by a critical evaluation of Lienert's (2010) study of the diffusion of Fiscal Responsibility Laws (FRLs). This writer uncontroversially traces this diffusion to the process that led to the implementation of the Fiscal Responsibility Act (FRA) of 1994 in New Zealand. He finds the innovative character of this reform in its being a limited scope law that 'elaborates on the rules and procedures relating to three budget principles: accountability, transparency and stability' (p. 5 original italics). In tracking the diffusion of this legislation to other countries, he identifies four core characteristics: (i) 'specification of the medium-term path of fiscal aggregates'; (ii) 'description of the medium-term and annual budget strategy for attaining the chosen fiscal objectives'; (iii) 'regular publication of reports (at least twice a year) on the attainment of fiscal objectives or targets'; (iv) 'audited annual financial statements that assure the integrity of fiscal information' (p. 5).

    These characteristics were most immediately reflected in Australia's 'Charter of Budget Honesty' (1996) and the UK's 'Code of Fiscal Stability' (1998) that eventually morphed into the Fiscal Responsibility Act of 2010. They also came to be incorporated in FRLs introduced in clusters of Latin American (Colombia 1997, Argentina 1999, Peru 1999, Brazil 2000, Ecuador, 2002, Mexico 2007 and Panama 2008) South Asian countries (India 2003, Sri Lanka 2003 and Pakistan 2005) and African (Tanzania 2001, Mauritius 2006 and Nigeria 2007) countries. Moreover FRL-type measures (incorporating some but not all of the four core characteristics) have spread even more widely to around 40 percent of emerging and 20 percent of advanced countries (Lienert, 2010, p. 4).

    The diffusion of FRLs thus seems to exhibit a number of the empirical characteristics that Weyland (2005) observes in other policy diffusion processes. These are: (i) 'commonality in diversity' with similar reforms being adopted in countries at different stages of development with different political and economic structures; (ii) 'geographic clustering'; and (iii) an 'S-curve' pattern of adoption, particularly within particular geographic clusters. Ultimately, the diffusion of FRLs seems to be mainly bounded by national and supra-national conventions pertaining to fiscal responsibility that function as effective substitutes for this legislation particularly in advanced countries or regions (such as Europe) where they have been institutionalized.

    Although Lienert (2010) does not offer a causal explanation for this diffusion pattern, this is provided in Weyland's (2005) survey that favorably compares bounded rationality with external pressure, mimetic and rational learning models of reform diffusion. In essence, bounded rationality explanations are based on the proposition that limitations in informational processing capacities induce policy-makers to resort to inferential short-cuts such as availability, anchoring and representativeness heuristics.

    Thus the availability heuristic, that comes into play when decision-makers focus on the most striking and immediate information would better explain observed geographic clustering in Antipodean, Latin American and South Asian countries better than other models that would predict a more geographically dispersed pattern of transfer. Also the tendency of boundedly rational decision-makers to anchor the process of considering other options so that they devote less time and resources to exploring adaptation possibilities would explain the substantive similarity of FRLs within these clusters compared to the rational learning model that would predict a greater degree of adaptation. Finally, the 'the representativeness heuristic' that explains how a wave of enthusiasm that 'induces people to draw excessively clear, confident, and firm inferences from a precarious base of data' (Weyland, 2005, p. 284) would explain the observed S-shaped temporal pattern of adoption in these clusters better than either external pressure or mimetic models that would predict an explosive global pattern or the rational learning approach 'that requires a careful cost-benefit analysis that considers a longer track record' (p. 280).

    This explanation of policy diffusion becomes more problematic as soon as questions are raised about exactly what it was about New Zealand's FRA that policy actors in other countries found innovative. In this paper we will argue that the New Zealand model has emerged as one of two models that countries currently seeking to adopt FRLs can follow. Thus while Australia and the UK appeared to have closely followed the New Zealand model with their respective adoption of the 'Charter of Budget Honesty' in 1996) and the 'Code of Fiscal Stability' in 1998 (see Table 1), this model can be distinguished from that adopted in the Latin American and South Asian clusters according to its commitment to qualitative principles of fiscal 'prudence' (New Zealand), 'honesty' (Australia) and 'stability' (the UK) rather than quantitative fiscal targets.

    The diffusion pattern of the New Zealand policy innovation thus differs significantly from that hypothesized by Weyland. Initially it appears to have been limited to a cluster characterized not by geographical but 'socio-cultural proximity' (Rose, 1993) created through the relatively dense linkages and sharing of ideas between policy elites. More interestingly, there has a recent upsurge in interest with governments in Kenya (2005), Nigeria (2006) and Mauritius (2006) pushing for legislation along New Zealand lines, while others such as Ireland are giving the New Zealand model a 'second look' after quantitative fiscal targets (such as those imposed by the 1997 European Union Stability and Growth Pact) have become difficult to sustain in the face of the global downturn and sharp reversal in economic and fiscal performance that many countries experienced following the financial crash of 2008 (Wallis, 2011). Policy innovations such as the FRA thus have the potential to break out with greater lags than would be predicted by bounded rationality models according to which waves of enthusiasm for them would tend to wane.

    In this paper we argue that these puzzling features of FRA diffusion should prompt a re-evaluation of the social constructionist perspective on policy diffusion. By equating this with a simple mimetic diffusion model, Weyland (2005) does not do it full justice. For example, he ignores the emphasis social constructionists such as Haas (1989), Sikkink (1993), Strang and Meyer (1993) and Ramirez et al. (1997) have given to the role geographically or socio-culturally proximate networks have played in the clustered diffusion of policy innovations.

    A more careful comparison between such social constructionist and alternative 'coercion, competition and learning' based explanations is provided by Dobbin et al. (2007). While these reviewers point out that 'the paradigms have developed independently', they helpfully delineate the types of innovation with respect to which particular models may be provide more persuasive explanations of diffusion. In sum, the most persuasive explanations of diffusion are provided by: (i) the competitive model with respect to reforms such as capital account liberalization and corporate tax breaks that reduce the cost of doing business in competing countries (Rodrik, 1997; Simmons and Elkins, 2004); (ii) the coercive model with respect to situations where either hegemonic countries or international institutions can muster the hard or soft power to impose solutions on 'weaker' countries (Mosley, 1995; Hira, 1998; Gruber, 2000); and (iii) learning models with respect to innovations such as privatization (Ramamurti, 1999) and pension reform (Brooks, 2005) where complicated implementation processes with the risk of perverse consequences may make countries wary to adopt them until they can draw conclusions on the basis of evidence generated by policy experiments elsewhere.

    By contrast, social constructionism seems to provide a compelling explanation of the diffusion of ideas-based norms or conventions in areas such as education (Meyer et al., 1997) or human rights policy (Ramirez and McEnealey, 1997; Boyle and Preves, 2000) where adoption confers international legitimacy on the countries concerned. It directs attention to the socially constructed nature of these norms and the mechanisms by which they achieve social acceptance--for example, through the example of leading countries (Haveman, 1993), the endorsement of epistemic communities of policy experts (Strang and Meyer, 1993; Kogut and MacPherson, 2007) or positive evaluations by peers who occupy 'structurally equivalent' positions in policy networks (Elkins et al., 2006).

    From this perspective, social constructionism would seem to be suited to the explanation of the diffusion of the type of qualitatively based fiscal norms embodied in New Zealand's FRL. This paper will try to make it clear how the ideas on which they are based are socially constructed and depend on internal legitimation by a broad-based consensus that must to some degree be mirrored in the countries to which...

To continue reading