This week Andrea Black (ex Inland Revenue and Treasury and former independent advisor to the Tax Working Group) and I take a closer look at the impact for Inland Revenue of its Business Transformation programme and find some puzzling discrepancies


By Terry Baucher

I’ve previously discussed on a number of occasions the impact of Inland Revenue’s Business Transformation on taxpayers and tax agents. This week we’re going to take a closer look at how Business Transformation has affected Inland Revenue itself. And to discuss this with me, my guest this week is Andrea Black, who when we spoke last year, was then the independent advisor to the Tax Working Group.

Prior to her role with the Tax Working Group, Andrea was at Inland Revenue between 2000 and 2012. During that time, she was involved in the Inland Revenue investigation of the Australian owned banks’ structured finance tax avoidance schemes. This ultimately resulted in the banks paying over one billion dollars in tax and interest.

In 2012 Andrea was seconded to Treasury, a move which then became permanent and she stayed at Treasury until 2015. She then returned to Inland Revenue for a final year.  Andrea is currently the policy director and economist at the New Zealand Council of Trade Unions. This is a role in which she has been working with the Public Service Association who have been supporting Inland Revenue staff during recent restructures

Morena. Andrea, welcome to the show. So, when you were at Treasury and Inland Revenue what was the expectation about Inland Revenue’s Business Transformation?

Andrea Black

The kind of vibe at the time was that it was going to become a knowledge-based agency, smaller, smarter. We could see that our colleagues in processing were likely to lose their jobs, but it looked like they were going to be good processes to support them through that and generally restructure the department.


Was the expectation at that time (2016) that Inland Revenue would lose one third of its staff?


To be fair, there was an expectation that there was going to be quite a lot of job losses. But if you thought about it logically, if you were going to have a much flasher computer, you were possibly going to need fewer people. And the old computer system FIRST was honestly held together by bits of wire and there were lots of people doing lots of things to try and hold it all together. So, you could totally see that when things were finished you might not need as many people.


Listeners may find this a bit surprising, but it’s actually quite unusual to sit in an Inland Revenue office and see the terminals they’re using. The only time I’ve done this was way back in 2012 when we were trying to resolve a matter involving transfers of imputation credits. I was shocked at what I saw, because the computer screen I was looking at was something from the 1980s.

The upgrade was well overdue. And what I was hearing was that there was something like 800 people whose role was to reinput data received from tax agents and taxpayers, so it could be serviceable and used elsewhere.


You know, that’s right, and then they did these dumps into something called Data Warehouse, which was how the analysis was done. So the High Wealth Individuals work that I led was all done from Data Warehouse. So in my new world there would be a greater use of capital, the labour would become more productive. It was not a surprise, you know, it was an expectation.


But you have been surprised by what’s been happening with staffing, not so much the numbers of staff, but the type of staff who are leaving.  This caught our eye as well, by the way.


Yes. We expected that our processing colleagues and some of our colleagues that were holding things together with bits of wire would find themselves without work. And we assumed that there would be processes for dealing with that. We also heard that they were going to have fewer layers and fewer managers. So, the expectation was there would be the types of changes that would be very orthodox and very ordinary. But then there were the changes in May 2017 after I left Inland Revenue.

I started hearing from so many unhappy colleagues. Processing staff seemed to be largely unchanged. But the cuts were falling on the senior technical staff, you know, the senior investigators and the principal advisors, the group I had worked with on the bank tax avoidance cases, and had then gone on to the Frucor case. These senior investigators had their pay being cut and the principal advisers were going to get mixed up with everyone else as technical specialists.

Now, one thing I just want to address is that I’ve heard it said that most people haven’t had a pay cut, and that’s possibly true. But it was the people at the top who were the most impacted. There’s a few things that’s happened. There are very able people who were managers and team leaders who come back as technical specialists. So that’s kind of a plus in the senior investigator and technical space or the principal adviser cohort. On the other hand, I would say possibly half of the senior top talent has left. That’s a lot of talent to lose.

I mean, and I’m sure you’re aware, there was 18 months where my LinkedIn feed was just going ping, ping, ping, with all  my long term friends and colleagues from Inland Revenue, posting that they were now working for themselves or were ow with another agency or now with a big four now or overseas. You would normally have expected a degree of turnover, but it was almost like every day or every well, let’s say every week, I don’t want to exaggerate. And yet  there was another one of these people moving on.


There’s something else that I think is an important point to be made here – that Inland Revenue salaries at that level you’re talking about are actually quite competitive because the Big Four accountancy firms are after people like yourselves and other policy advisors, who are highly experienced people and technically very good. So Inland Revenue has to be competitive.


It might not necessarily be the end of the world, because there were really capable technical people who had become team leaders and managers, who came back as technical specialists, so with the remaining people I wouldn’t necessarily be worried or concerned. But there has also been a massive contraction in morale, which I think you’ve alluded to, Terry, in the past with the engagement scores.


Yes, I have been tracking Inland Revenue staff engagement scores. In Inland Revenue’s report to June 2019, it was at 29%, which unbelievably was actually up from the previous year.


I did a contract with an agency last year they were at 60% and wanted it to be higher. At my time at Treasury, I think 60% would have been really low.


So Inland Revenue has a problem. Now you’re supporting the Public Service Association.


Yes, when I started working with the PSA one thing came to mind, I think it was last year following the Commissioner’s presentation to the Finance and Expenditure Committee. She was being queried about the number of audit hours. She said yes there had been a fall in audit hours, but Inland Revenue was still meeting its return targets. And I remember thinking at the time  “Wow, that’s really impressive.”

But later I got thinking and said to the PSA whether it would be worth finding out about the hours, but also the composition of Inland Revenue’s return.

There’s a lot of talk about how Inland Revenue get seven dollars return for every dollar it receives in audit. And that’s right to a point. But that amount is an identified “discrepancy.” And this discrepancy is a change in tax position.

And if we look at the definition of “tax position” in section 3 of the Tax Administration Act 1994, pretty much anything, any interaction between Inland Revenue and taxpayers, is a tax position.

So a tax position includes, say, the tax avoidance cases with the banks, or a case where someone says they earned nothing but Inland Revenue finds they did earn a million dollars. The tax on a million dollars would be a discrepancy or change in the tax position.  Both of those involve a liability for an amount of tax and these are matters people would consider a discrepancy.

But a discrepancy is also a change in a memorandum account, or a change in the amount of tax losses. So if Inland Revenue finds that someone had overstated the losses, that counts as a change in tax position. And all these things are what I would call good work. All of these things are important activities in the investigation function.

But if you think about it, there’s different degrees of effort and different degrees of pushback that you get from a taxpayer. There’s a spectrum. So on one hand, you would have changing someone’s tax return which means they have to pay more tax and they really don’t want that.

But on the other hand, you could have an adjustment of losses for a company that’s perennially a loss maker or, and this is my favourite example, adjusting the imputation accounts of New Zealand subsidiaries of Australian companies. After the double tax agreement with Australia changed, you no longer had to attach imputation credits in order to be exempt from non-resident withholding tax.

In those circumstances, changes to losses or imputation credit accounts aren’t met with much pushback because they don’t really affect the material position. It’s still useful work but it’s a sort of lower quality discrepancy versus the higher ones where you would be changing someone’s taxable income in a way that they had to pay tax.

Now, from an investigation perspective, the top end, like in the bank tax avoidance cases, you know, taxpayers throw everything they’ve got at it in resisting amendments and it’s a lot of work. And so, a discrepancy coming from that is really hard won.

All the time that I was in the field, I saw all the effort really going into what I call the high value discrepancy. I mean, there would have been some other stuff that went on. You know, like if you’re doing an audit, you find things. But I found that the audit programme was really targeting only what I call the high quality discrepancies.


And speaking from the other side, those were the fights that we’d be having with Inland Revenue: what was deductible, what was taxable. We’d sort out arguments about incorrect imputation credit accounts and losses brought forward relatively quickly and resolve those issues. And we don’t tend to dig into the trenches for them. But the other point of these highly more technical matters, yep, that’s where the fights happen.


That’s right. And so while strictly speaking, everything is a discrepancy, everything I always saw when I was in the field at Inland Revenue was the audit programme was targeted to the high quality ones, but low quality ones might just tick in.

So knowing that, when I was hearing that even though the audit hours had gone down, Inland Revenue was still meeting its targets, I was like “Okay. There’s a couple of ways that could happen. The new computer could be really, really flash and be doing an awesome job. Or maybe there’s another option.”

So, after discussing it with the PSA I filed an Official Information Act request asking for audit hours by year and broken down by area. I also asked for discrepancies split by type. And I gave the examples of loss adjustments and imputation adjustments. Because I know in the past investigators always coded those things up.

So if we saw, like, lots of going after people’s income tax. I’d have been really comfortable; the Department would have been just awesome to be able to deliver that with really few audit hours.


Yes. The Official Information Act (OIA) request that you got back is really very interesting reading.


So in terms of audit hours, we’ve got a breakdown for each of the years ending 30th June 2016 which would have been the last year I was at Inland Revenue, through to 30th June 2020. And this is a version of what came from the Andrew Bayly Parliamentary questions from last year.

And I think the PSA is looking to make the OIA public. But in the meantime, the PSA is happy for me to send it out to anyone who wants it.


So total audit hours have fallen by two thirds roughly and hours on complex technical issues are down 90%.


That’s right. That’s right. So I would have expected the discrepancies would have similarly fallen, but maybe that’s old thinking. The total has gone from $1.2 billion to $958 million, which is less but not a lot less.

So complex technical I found fascinating in terms of the discrepancy. It’s gone in the 2015/16 year, which was when I was there, from $296 million in discrepancies. It peaks in 2016/17 at $462 million and then it goes to $295 for the latest year.

So on the face of it, yes absolutely, the Commissioner is right. While the audit hours have gone down the return or the discrepancy is unchanged, which is why I was really interested in what the discrepancies represent because, one can think thoughts, but it’s important to see what the facts are. And like I say, I knew that the investigators always coded this up. But it seems like the new system doesn’t tell you how it’s broken down.


A little bit aside here this is a point I’ve had an interest/concern about for some time is that Inland Revenue has access to more data relating to the state of the New Zealand economy than practically any other organisation, but we don’t see a lot of it.

I’m always disappointed how little data Inland Revenue releases. Because my view, and you may well share this, the more you tell people about what you know, the better enforcement is, because people aren’t going to try fancy stuff.

Inland Revenue will know for example how much is being spent by the average bakery, and what proportion is cash, what proportion is EFTPOS. They know that all those details.  So people shouldn’t try and hide cash sales because it’s not going to work. That sort of detailed information Inland Revenue could be doing more with, but isn’t doing so at the moment.

So to find out that the system doesn’t appear to carry that level of detail, I have major concerns. Not just as a tax agent, but also as a taxpayer who’s had to pay a substantial amount of money – $1.2 billion I think – on Business Transformation, one of the biggest I.T. upgrades in the country.

And you’re telling me Inland Revenue actually aren’t able to draw more data? Because I know there’s a lot of very discontented software providers in New Zealand who didn’t get a chance to deal with this project.


Here’s what Inland Revenue said in the OIA.

“You have requested the total discrepancies be further broken down by time. I’m only able to provide loss reduction adjustments in part where these have been captured manually for the reporting by staff in our new system (START).

“The balance of any loss reduction adjustments and all imputation adjustments are either recorded in one of our heritage systems (eCase) which has been decommissioned, or as other information in START. To provide these figures, would require a manual review of individual case records.  Accordingly. I must refuse this part of your request…because the information sought cannot be made available without substantial collation or research.

It astounds me because what they are saying even is that even the managers don’t know how these are broken down. I’m speechless, I don’t know where to start with it.


Wow, just talking generally about Business Transformation, obviously, I’ve talked in the past about how we as tax agents feel it’s gone.  To say underwhelmed would probably be unfair, but we have not been happy with some of the side effects of it.

So are you telling us Inland Revenue have got a system that doesn’t work well with agents who are the principle people who work with it and also doesn’t record information regarding audits?

Another thing is we know is that if they want to make any proprietary adjustments to the system, they have to get sign off from all the other tax authorities that use the same system. There are six or seven others, I think, including Finland. So there’s quite a lot of questions around how has been sold to us and how it’s actually operating.

And then come back to the point you started with – there’s been a fallout with low morale in Inland Revenue. That is not good for the tax system, let alone for Inland Revenue itself, but for us as taxpayers. I don’t think it’s a good system when the person on the other end of the phone is less than happy to be there. What is the PSA finding on this stage? It’s not particularly encouraging from what you have said.


The PSA is in the middle of some processes so I don’t feel I can speak for them. They are looking at supporting staff through this, is probably the short answer to that question.


A very diplomatic answer.  So what you’re telling me is a little disconcerting, because it sounds like these numbers may not mean everything you’d expect them to mean.  And talking about audit activity, we’re not seeing as much audit activity as previously. I think Covid-19 might have caused some interruption, but there definitely seems to be lower audit activity going on.

And I have to say, I do struggle with the numbers here which suggest Inland Revenue is identifying roughly the same amount of discrepancies but with one third of the hours. If the new system is as efficient as that, then they should be exporting it. The productivity gains are enormous. The New Zealand economy would go gangbusters if we could get those productivity gains across the board.

We’ve got an overall decline in discrepancies as well. The amount of revenue taken on this has declined from one point three billion in the year to June 2017, to now 958. And it’s a downward trend.

And they’re not seeing much gain in the hidden economy either. That number is constant roughly as well, which again surprised me because we know there’s a lot more out there. So what’s your overall view of what’s going on here?  Because when you drill into it, it doesn’t seem to stack up quite as well, does it?


Well, to be fair, when I looked at the Inland Revenue funding for an article I wrote a few weeks ago, it’s quite interesting to see that investigations funding is falling.


What? Did you say investigations funding is falling?


Yes, funding for investigations and debt management funding is falling, but you would expect Inland Revenue funding to fall overall because of all the money that’s gone into Business Transformation. But interestingly, investigations and debt management funding has fallen but processing has increased, which I don’t understand at all, because the new computer system was supposed to make processing much easier.


Yes, I agree that doesn’t compute because if you’re designing a smarter system, you’d be putting the resources to where the smarts are, i.e. investigators. You’d also be wanting to put resources into debt management because it has been a longstanding problem for Inland Revenue Which to be fair to them, I think they’re now getting on top of. But now it’s got to deal with Covid-19 as well. When you think about it, Inland Revenue has got the Covid-19 fallout, a historic debt problem, so against that background falling debt management funding does not compute.


I think it can be quite computable because if you have this idea that you’ve got a much bigger and flashier system that can take away the lower level work for debt and investigations, you could see that funding in those areas funding would fall. I always want more money in investigations, but that has a degree of logic to it. The logic of processing funding increasing, I can’t get my head around.


What is the scale of the increase in processing funding?


It’s gone from about $100 million to about $150 million.  Debt funding has gone from about $150 million down to under $100 million.  Investigations has gone from about $175 million down to just over $100 million.


On debt management, I think it’s possible Inland Revenue can get certain big efficiencies if it gets into the act of chasing debt earlier.  If you’re not leaving it for years before you start asking questions, you will get some easy productivity gains.

But we’re always hearing, as you mentioned at the beginning, that investigations generally bring in seven dollars per dollar invested return. But at a time when the government’s books have been shot to bits, we are hearing these numbers are pointing to a reduction in the most valuable return the government can get from its main source. Yet there’s a 50% increase in processing funding for a new computer system on which we’ve already spent $1.2 billion dollars already.

I would want people providing oversight such as the Finance and Expenditure Committee to be asking a few more questions about this, because it does not stack up from where I’m sitting.


Well, I just want to say one thing. We don’t necessarily know how valuable that seven dollars return we’ve been talking about is. As we’ve discussed it’s a tax position but it’s not money in the bank, it’s the amount before collection. But the other thing I would say is that debt has been rising since 2017. Looking at Inland Revenue’s June 2019 annual report, it’s gone from $2.9 billion, to $3.1 billion in 2018 and $3.5 billion in 2019.

So, I would agree with your comment. I found this out when I was preparing the Official Information Act request on behalf of the PSA and I was just, quite frankly, astounded.  Although I don’t like the fall in funding for audit investigations and debt, as a former Treasury official, I can see the logic of that, given all the capital expenditure that has gone into Business Transformation. But I was just floored that funding for processing obligations and entitlements has increased.


Well, that’s quite a few bombshells just dropped in there about what’s actually going on. We have a new Minister of Revenue, David Parker, and a Parliamentary Under-secretary, Deborah Russell as well. They will get what’s known as a Briefing to Incoming Minister which will be released later this year in due course. Now, it’s going to be very interesting reading to see whether some of the issues we’ve just been talking about actually get picked up. But in the meantime, I would hope that other oversight bodies, such as the Finance Expenditure Committee, start asking some questions because it’s quite concerning that there’s a discrepancy, to use a word, between what’s being reported and what actually might be happening.

Well, I think we’ll leave it there. Andrea, thank you so much for coming on and great to have you on the show again. That’s it for this week. Thank you for listening.

I’m Terry Baucher and you can find this podcast on my website, or wherever you get your podcasts, please send me your feedback and tell your friends and clients. Until next week, ka kite āno.