discussion thread focusing on the complexity of incorporating census data
in SD models and the general utility of data-intensive modeling. These
questions do not deal with "technical" aspects of SD modeling, but refer
to the potential for application, and the significance of using, SD in
public policy development.
In his posting of 2/8/00, Keith Linard wrote:
"Sometimes, particularly in circumstances where solutions are non
controvertial or where there are not multiple stakeholders pushing
conflicting agendas, the successful model may simply consist of causal
loop or influence diagrams , or primitive SD models. Much of the public
sector reform program in Australia during the 1980s was driven by such
high level modeling."
1) I have searched the listserv archives extensively for messages
relating to public sector-related applications of SD, as that is my
primary area of interest. (Jaideep, take note. Thank-you for this
wonderful service!
I do not recall finding any mention of a published article canvassing the Australian experience in using SD for
public sector reforms. I am aware of the articles presented at the 1999
SDS Conference in Wellington, several of which deal with specific public
sector applications in a variety of countries. For more than a decade,
we in Canada, have looked to Australia (and New Zealand) as true
innovators in public sector reform. I would be greatly interested in
learning about SDs role in this area. Are there any published articles
on the topic?
2) Keith Linard also wrote in the same posting:
"Government has a variety of possible policy levers to influence
change.......... A critical issue for all will be how different policies
impact on their particular clientele compared with other winners and
losers."
(Note: I have deleted the text that is specific to the vehicle emission
example he was citing.)
I have been pondering, for some time, whether SD offers a superior
alternative to the standard approach adopted for "regulatory impact
analysis" which, at least in Canada, is centred on cost-benefit analysis.
The first step in regulatory impact analysis is to properly frame and
understand the problem. Clearly SD offers great potential at this stage.
The second step, after deciding whether government should intervene, is
to identify and assess alternate courses of action: alternate
instruments (policy levers). In the Canadian federal government, this
stage goes by a variety of names: "regulatory alternatives analysis",
"strategic options analysis", and most recently, "risk management
analysis". Seems to me that if a SD model reflects a comprehensive
theory of what is influencing the (problem) behaviour, then the model
could be used to help identify and assess possible consequences of using
different alternatives. I have not been able to find any instance of its
use in Canada for that purpose, however.
The third step is to assess the impacts (largely focused on costs vs
benefits, and distributive impacts (winners and losers) of the proposed
regulatory approach. This can include an assessment of the potential
impacts of regulatory requirements of different stringency. There, it
seems that application of sensitivity analysis for some variables in an
SD model could help inform the analysis. (I think that the model would
have to incorporate a "theory of compliance" more sophisticated than that
which public choice theorists would advocate.)
So far, all this adds up to the idea that SD would be a useful addition
to the regulatory impact analysis toolkit. What is troubling me is that
I think it probably offers something much more significant, something
akin to a "paradigm shift". The genesis of regulatory impact analysis
methodology is microeconomic analysis. The frame of reference for the
Canadian federal regulatory policy (ie. the policy defining when and how
the government will use regulatory intervention) is narrowly bounded by
economic principles (particularly cost-benefit). But, after 20 years of
living under this regime, it is clear that it doesnt fully connect with
the considerations that drive public policy development and
decision-making at the political level. And so, it has not been fully
successful in shaping government decisions. Perhaps, not successful at
all! (Fixes that fail!)
My problem is that I cant quite get my thinking straight on what it
would mean to adopt SD as a way to frame regulatory impact analysis. Is
it simply a better tool, a way to take account of a broader range of
considerations and make sense of their interplay over time? Or is it
something much more fundamental, equivalent to the revolution that
occurred when governments began to understand that regulatory
intervention created economic impacts outside government that
significantly outweighed the budgetary impacts inside government? I have
no doubt that it is a valuable addition to the toolkit. My instinct
tells me that it is something more.
I would appreciate any input from the Discussion Group that could help
advance my thinking in this area.
Eric Milligan
From: milligan <milligan@rcgi.com>
RCGI
The Regulatory Consulting Group Inc.
Suite 600, 45 Rideau St.
Ottawa, Ontario
Canada
K1N 5W8
Tel: 613-562-4077
Fax: 613-562-4102