Rachel Siewert (WA, Australian Greens) Share this | Link to this | Hansard source

I rise tonight to speak on the Cashless debit card trial progress report, which was released on the final day of last month, 31 October 2016. I would like to point out that in the most recent estimates I asked the Department of Social Services about a so-called confidential report that had been quoted in the media on the progress of the trial sites. I asked about this report because over the course of the last few months we have seen the Minister for Human Services, Mr Tudge, repeatedly drip-feed mostly anecdotal feedback to the media as a means of promoting the cashless welfare card trials. Then he started quoting some figures. Mostly his spin on the figures was, of course, around the positive side of the card. When I asked the department to table the results—this was during estimates—that the minister and others had been quoting, including in the Press Club, they were unable to provide those results. They said they did not have the results on them. When I asked about whether there was a confidential report, they said, 'No, there was not a confidential report,' and they were unable to provide the detailed figures the minister had been quoting. The department took on notice to look at what information was available. However, a week and a half later the department suddenly released the progress report that supposedly outlines the progress on the card. I found the turnaround of this report remarkably quick, considering the department was not able to provide any details and in fact said during estimates that there was no report. I am astonished that a week and a half later they could suddenly come up with a report and have it published. I must admit this appears to be very convenient and raises a very big red flag.

The report is basically not worth the paper it is written on. It contains a combination of very premature analysis of very early data, small datasets, ad hoc datasets and unreliable anecdotal feedback. It does not take in any other current programs in place or external factors that may influence some of those data sets. The report does not produce credible or substantive evidence that the card is working. For example, one of the many failures of the report is that in a small dataset in the Kununurra trial, where the WA police reported 29 assaults for April 2016 and a rise to 59 assaults in September. I note that for this statistic there is no comparison with previous, unlike other statistics in the report. These assaults are attributed to an influx of tourists. Some people could just as well say, 'Well, the card isn't working,' but, of course, it is too early. The dataset is quite small and we suggest it needs a proper analysis. In using this excuse when it suits them, the government is acknowledging that in some instances there are external factors that may result in the indicators that they are seeing, but they only do that in certain circumstances.

Another example of external factors and wider context is the previous data that indicates that gambling in Ceduna, or in the region in which Ceduna is in, has been on a downward trend for the last five years. Given that the cashless welfare card recipients comprise only 30 per cent of the population in the area, it is impossible to isolate this group and draw genuine evidence about the correlation between the card and gambling, particularly when the data is about the use of pokies, not just in Ceduna but in other regional towns where the trial is, in fact, not being rolled out.

External factors that might influence measurable indicators is something that I have mentioned repeatedly since the trials were legislated last year. I have also repeatedly said that the government should be funding services in a similar trial site that does not have the cashless welfare card so you can isolate, or are more likely to be able to isolate, the potential impact of the card. The government refused to do this, and so we will not know whether some of these improvements, if they are genuine improvements over the long term, can be contributed to the card or to the wraparound services that are being injected into the trial communities.

I read in the report that the admissions to the Wyndham Sobering-up Centre dropped 36 per cent compared to the previous year, which, you have got to say, is good news. But if you flick three pages forward in the report you see that there is an injection of drug and alcohol support workers, youth and family support services, an adult rehab program and an alcohol and other drugs brokerage fund. That is good news. Those are the sorts of services that the community needs, but how do we know whether it was the card that was responsible for the decrease or those excellent services that should have always been provided to that community?

This is money well spent, and I do support the government providing these services. But, as I said, we will not know whether it is these services that are producing that current decline, bearing in mind this is very early data. The government has created a so-called trial where it is nearly impossible to genuinely measure the success of the card, because there is nothing to compare it to in terms of those services. In saying this, I can 100 per cent predict that the government will attribute the success to the card no matter what. That is obvious from the selective use of data that is contained in this report.

The government has done nothing but use anecdotal feedback. It has also used anecdotal feedback to suggest that the trial is successful. Unsurprisingly, a lot of this anecdotal evidence was also included in the report. They have been continually rolling this out for a couple of months. The report is clearly a political puff piece when anecdotal 'evidence' includes a line from the extremely pro cashless welfare card Mayor of Ceduna saying that the card 'is the best thing that has ever happened.' How could you include this sort of comment in a genuine progress report?

According to the report, 'promising anecdotal evidence' in Kununurra and Wyndham includes 'people not buying playing cards'. For goodness sake! Then, of course, in small print down the bottom, the report says:

… All anecdotal statements are individual opinions or unverified—

unverified!—

data sets. The underlying cause of each claim has not been independently verified, tested for statistical significance, or placed within its wider context.

Yes, that is exactly right. It is anecdotal evidence that has not been provided in the wider context, yet the minister and the government rolls it out as evidence that the card is working. On top of this the anecdotal data does not match up with the datasets. It is a debacle. In fact, when you looked at the Northern Territory intervention and talked to people about whether things were better or not, some people said, 'Yeah, they are better', but, when you looked at the data, the data showed that things were not better. So the anecdotal evidence is not being backed up by the datasets.

Obviously the report has not looked at some of the police figures that are available for South Australia, and I urge the minister to look at some of those police reports. I am not saying this should be used to evaluate this card's success or not, just as the government should not be using the data that they are collecting at the moment to talk about the success of the card and use it out of context without looking at the external factors. But perhaps the minister should look at the data from the South Australian police report that shows in the Eyre Western LSA an increase in criminal activity from the previous year. This includes robberies and related offences, up 82 per cent; aggravated sexual assault, up five per cent; and serious residential criminal trespass, up 44 per cent. That is not mentioned in the so-called progress report—although some of the graphs in the progress report do show a slight dip straight after the card came in, but those stats are going right back up again. This is not an appropriate way to report on a very serious trial that the government is carrying out.

We do not support the trial—I have that on record well and truly—but if the government was very genuine about this so-called system, they would not be doing this sort of anecdotal rollout of talking to somebody when they happen to run into somebody in the street, or misuse datasets, or misquote datasets and use them out of context without actually putting in the results of the external factors. There is quite a bit of evidence from other measures like this that have been successful at the beginning and then, of course, as people got around them, they dropped off. That is what is going to happen here.