Monday, May 04, 2009

Monitoring the 90-day bill

The Department of Labour plans to fulfil Kate Wilkinson's promise to monitor the 90-day fire-at-will law by tracking how many people look at its website and how many people call its call-centre to complain.

The monitoring plan is contained in a memorandum, Monitoring and evaluation of the introduction of a trial period into the Employment Relations Act 2000 [PDF], which I obtained using the Official Information Act. I also obtained several early drafts of the memo, and email correspondence about its development.

The memo notes that there are "significant gaps in the data" which make it "difficult to assess whether the intent of the policy is being met". If anything, this understates the case. An appendix to the document lists the detailed research questions around the introduction and ongoing effect of the law, and the available data which could answer them. There is "no specific data" on whether the law has benefited first-time workers, Maori or Pacific Peoples. At present, they have no way of measuring how many people have been forced into such agreements, let alone how many have been laid off. They have no way of measuring whether it has led to increased employment and whether that employment is probationary or not (they propose a possible data source, but it is very noisy and will not include data on probationary periods). They may be able to find out if it leads to more beneficiaries being employed, but only if WINZ tells them. And their plan for measuring the impact on "employment relationship problems" is to look at website traffic, call-centre statistics, and complaints to mediators.

And that's the good version. The development emails paint an even worse picture. While the final memo talks of using mediators as a data source, the policy analysts pointed out that the ability of employers to dismiss workers without cause during a probationary period meant that such workers were unlikely to end up before the mediation service, while the confidentiality of the mediation process meant that mediators would not record the cause of the dispute anyway.

The memo also suggests that more robust information could be gained around the end of the year by using a Stats Department database called LEED (Linked Employer-Employee Data). But an email from the stats department [PDF; some extraneous material at top of page] showed that that was unlikely to be effective, with any change in the stock of short-term jobs lost in the noise of recession, and no ability to identify individuals fired under the bill. This means it would be

difficult to evaluate the outcomes for employers and employees who are affected by / make use of the 90 day bill.
As with the mediation issues, these concerns seem to have been ignored. As a result, the Department's "monitoring plan" basically boils down to web hits and false hope.

The memo notes that the lack of data "poses risk to the Department from not being able to effective[ly] measure what is occurring" and that it will not be able to answer questions on the effects of the policy and whether it is meeting its objectives. And it's not even interested in tracking abuse. So much for Wilkinson's promise...