The ANAO Performance Audit of the Community Development Program

The ANAO released a performance audit of the Community Development Program this week (link here). The Community Development Program applies to most income support recipients resident in remote Australia, and 80 percent of its 33,000 participants are Indigenous. The audit was quite narrow in its aims, since it was designed to primarily assess the effectiveness of the transition from the Remote Jobs and Communities Program (RJCP) which was established by the previous Government in 2013 and ran through to 2015. Nevertheless, the ANAO report outlines a wealth of policy relevant information and for those interested in remote Indigenous policy, repays close reading.

I previously commented on the CDP program in a March 2016 post titled ‘The worsening crisis in remote employment programs’ (link here) where I reported on academic research raising serious issues with the programs breach rates and discussed the perverse incentives embedded in the scheme. The program is currently the subject of a Senate Inquiry which is due to report on 15 November 2017 (link here), although there was some commentary in Senate Estimates this week that a request for an extension to December was being made.

The initial media response (link here paywall) to the audit report in The Australian (‘Dead used to inflate jobs numbers’) focussed on accountability issues amongst the providers delivering the scheme thus relegating the responsibility of the Department for the administration of the program to a second rung, with the Minister nowhere in sight.

This is not surprising as the ANAO appears to go out of its way to make its headline conclusions as anodyne and boring as possible. The single recommendation in the report, that:

 the Department of the Prime Minister and Cabinet review the Community Development Programme provider payment structure, particularly the incentives it creates and its alignment with the underlying policy objectives of the program changes  (para 2.27)

deals with only one quite obtuse aspect of the multiple findings in the body of the report, and crafts it in terms so general that the Department could theoretically agree to the recommendation even where it intended not to make any changes at all.

How else to interpret the focus on third party providers in the ANAO’s headline conclusions and single recommendation, and the concomitant diversion of focus from the role and contribution of the Department (and behind it the Minister), in relation to the Community Development Program?

This goes to more fundamental issues relating to the role and influence of the public sector’s oversight agencies. They invariably set out to avoid conflict, to persuade, and to nudge, but not to hold agencies (and their ministers) strictly to account. Real accountability has become too risky, a bridge too far. In theory the key oversight agencies report to the Parliament, however the reality is that the Executive dominates and controls most of what occurs in the Parliament, with the result that the potential freedom of action of oversight agencies has become circumscribed. Yet we continue to fool ourselves that our oversight agencies are independent. They have the potential and capacity for independent agency, but if they choose to flaunt it, or even exercise it freely, their freedom of action, their budgets, their access to the networks of influence would be swiftly circumscribed. One has only to look to the treatment meted out to the Office of the Australian Information Commission to see this in action (link here).This is an issue which extends well beyond the Indigenous policy issues which are the focus of this post; nevertheless, it deserves wider consideration and debate than currently occurs.

Notwithstanding the headline conclusion of the ANAO report: ‘The transition from the RJCP to the CDP was largely effective’ (para 4), the ANAO report lays out a myriad of issues which reflect poorly on the way the Minister (as the public official ultimately responsible) has administered the CDP. As ever, the devil is in the detail.

I won’t attempt to analyse each and every issue of concern raised by the ANAO, as I wish to focus on some higher level issues. Nevertheless, as a preliminary step it is worth pointing to a number of the deeply problematic issues identified by the ANAO in their detailed analysis. These issues include:

Ø  the design of the CDP was not informed entirely by sound analysis of the RJCP (para 5);

Ø  a series of problems with the incentive structure for providers (paras 4 and 6);

Ø  serious under-subscription of two complementary schemes, the Employer Incentive Fund and the Indigenous Enterprise Development Fund (para 9);

Ø  the transition implementation timeframe from RJCP to CDP was too short resulting in providers not understanding the new program well, and many of the implementation risks identified by PMC in its implementation plan coming to fruition (para 13);

Ø  there is scope for greater transparency in the Regional Employment Targets built into provider agreements (para 21), and

Ø  the phone contact call times have tripled in the two years since CDP was established (para 3.32).

In addition, and surprisingly given the Government’s rhetoric on strengthening evaluation, the ANAO reports that

Ø  the evaluation strategy was developed seven months late in early 2016, and

Ø  a summary was not approved by the Minister until April 2016 (ten months after commencement).

According to the ANAO, this reduced the scope to collect performance relevant data. Moreover, the evaluation strategy was not peer reviewed by an independent reference group, and evaluation strategy milestones were not aligned with the Government’s timeframes for considering future funding for the program (paras 23 and 24). It seems that in the case of the CDP, data is not gold, but dross.

Buried in the ANAO report are a number of significant accountability findings including

Ø  the identification of poor financial management of provider payments by DPMC in the transition phase including a finding that 30 percent of payments / commitments in 2014-15 were not supported by evidence and thus not payable (paras 2.37 to 2.54- in particular 2.41); and

Ø  poor provider performance including at least nine instances of potential fraud (paras 3.38 to 3.52  and table 3.5). This latter issue was the matter which drew the attention of The Australian in their reporting (link here) whereas THe Guardian focussed on breach rates before picking up providers counting the dead in its second paragraph (link here).

A further serious issue identified by ANAO was the substantial underspends in the program, amounting to around $50m in 2015-16 (para 3.13). The ANAO indicated that the reasons for the underspends relate to ‘fluctuations in the total number of jobseekers in the CDP and lower jobseeker attendance in Work for the Dole activities’. The ANAO also refer (in para 3.56) to a number of prepayments being linked to the existence of underspends, code for the department spending money in advance of need to avoid an even larger underspend.

While the ANAO make reference to the extraordinary rates of breaching elsewhere in the report (paras 3.28-29), they don’t draw the link here between tougher requirements and participant preparedness to vote with their feet and walk away. The propensity for this to occur can only be magnified in locations where income management or the cashless debit card operates. As pointed out in my previous blog post, the assumption by policymakers that incentives that gain traction in mainstream contexts will also gain traction in remote Indigenous contexts is highly problematic. The end result is an ever more punitive system which fails to deliver outcomes.

The bottom line however is that the underspend translates to $50m in welfare benefits which did not reach Indigenous income support recipients, who are amongst the most disadvantaged citizens in the nation, and apart from the pain inflicted on individuals has the additional consequence that the funds are not available for circulation within remote regions’ economies. Notwithstanding the Government’s stated support for economic development, policies which have the effect of limiting the quantum of entitlements transferred to remote citizens also limit economic activity within remote contexts. This is an inevitable consequence of the hard line ‘welfare reform’ policy changes being pursued by the Government. The substance of this point appears to be accepted by the Government as the DPMC submission to the current Senate Inquiry into the CDP referred to above notes that the Government:

is considering options in a new employment and participation model for how any loss of income support, due to the application of penalties, might be invested back into community (page 11).

Perhaps the most significant performance issue identified is the cost effectiveness of the program. In comparison to RJCP, CDP costs around twice as much per jobseeker, and costs around five time as much as Jobactive, the mainstream noon-remote employment program (paras 3.9-10).

Yet despite the doubling in unit costs, employment outcomes (as a percentage of jobseeker placements) were only marginally better under CDP than RJCP: see para 4.16 which notes that in relation to 13 week employment outcomes, RJCP averaged 41 percent while CDP averaged 47 percent. The rationale for the huge bureaucratic shift from RJCP to CDP appears virtually non-existent (notwithstanding the ANAO statement that is too early to assess whether CDP is administered efficiently (para 15).

But the reality is not in the percentages, but in the absolute numbers (for both schemes). There are 33,000 CDP participants across remote Australia, but employment outcomes were 254 per month for RJCP and 311 for CDP. That is, less than one percent of CDP participants are obtaining 13 week employment outcomes per month. On these metrics, both the RJCP and CDP objectives of ‘supporting jobseekers and reducing welfare dependence’ (para 1) appear laughably ineffective. And one might ask, at what financial cost to taxpayers? And at what human cost to remote citizens?

Yet the Minister is quoted in The Australian as welcoming the report, saying on the one hand it had recognised effective implementation of the CDP, but foreshadowing major change on the other:

We are getting jobseekers in remote Australia off welfare and into work….we will soon commence consultations on a new remote employment model for Australia.

The Minister introduced CDP a mere two years after the introduction of the RJCP in June 2013, citing increases in welfare dependence under the program, and within two years was announcing consultations would begin to devise yet another remote employment scheme. See the discussion in Senate Estimates this week (link here) which outlined early moves to begin consultations for a new remote employment programs based on wages rather than income support.

While the Minister and the Department claim that the decision to replace the RJCP was based on strong analysis (refer paras 2.2 to 2.7 of the ANAO report), the underlying confusion identified by the ANAO strongly suggests that the decision was about politics as much as policy. In particular, the ANAO note in para 2.7 that the DPMC provided advice to the Government in October 2014 that there had been a significant reduction of 90 percent in jobseekers achieving 26 week employment outcomes compared to the previous year’s outcomes. Yet when requested to provide evidence to support the advice, DPMC were unable to do so.

If the decision to establish a new program in 2015 was justified, and the new program is working as the Minister claims then there would be no need for further change. On the other hand, if it is not working (as the cost data, breach data, and underspends suggest), then the Government needs to take responsibility for the failure and fix it. The Minister cannot have it both ways.

In his recent Wentworth lecture, the Secretary of DPMC Dr Martin Parkinson, was critical of the constant change in the Indigenous affairs machinery of government. He pointed to the opportunity costs of fractured relationships between government and communities, constraints on knowledge transfer and capability which arise from what he termed ‘constant churn’. He could as easily have been discussion the recent history of remote employment programs.

At the moment, the Minister seems to be trying to walk on both sides of a barbed wire fence. No wonder he appears uncomfortable defending his record.