sign up log in
Want to go ad-free? Find out how, here.

Terry Baucher on the latest Inland Revenue guidance on non-resident employers, calculating foreign tax credits, and whether the proposed social insurance scheme could incentivise more honest tax filings

Business / analysis
Terry Baucher on the latest Inland Revenue guidance on non-resident employers, calculating foreign tax credits, and whether the proposed social insurance scheme could incentivise more honest tax filings
IRD with eye on remote working
Image sourced from

We're now in the third year of the pandemic and one of the things that's emerged is our working patterns have changed as many people are working remotely, not just from home, but in completely different countries.

This has prompted Inland Revenue to issue an operational statement setting out what would be the obligations for a non-resident employer in relation to pay as you earn, FBT, and employer superannuation contribution tax. This applies where the employer is based in, say, the United States, but the employees are working remotely here. A scenario that I've seen quite a number of times in the past couple of years.

Basically, the operational statement confirms that a non-resident employer will have an obligation to withhold PAYE on payments to an employee if the employer has made themselves subject to New Zealand tax law by having a sufficient presence in New Zealand and the services performed by the employee are properly attributable to the employer's presence in New Zealand.

What that means is if the employer has a trading presence in New Zealand, such as carrying on operations and employing a workforce, that's normally sufficient for it to have PAYE obligations. And that would be, for example, it has a branch or a permanent establishment. Something always to watch out for is someone is signing contracts in New Zealand and performing those contracts in New Zealand with employees based here for that purpose. But no PAYE obligation arises where the employee decides I'm going to work/ return to New Zealand because it's safer here, and I can work remotely. That case is not covered.

The paper gives a quick example of an architecture firm based in Boston and one of its employees, George lives in Wellington.  George participates in virtual meetings, complete all his work in Wellington. But because the Boston firm has no New Zealand clients, all the work is carried out relates to American work, and the company has no obligation to deduct PAYE.

But what about George's position here? And this is something that can slip through the details here. In some cases, he is treated as self-employed and pays provisional tax. That's not technically correct. In fact, what he should do is to register as what they call an IR 56 taxpayer, and he files the employment information and pays PAYE to Inland Revenue. So literally he accounts for his own PAYE. Alternatively, the Boston firm could register as an employer and make the deductions on his behalf.

This is a scenario we've seen increasingly, so it's useful to have some guidance from Inland Revenue in the form of this Operational Statement. It’s also worth pointing out there is also sometimes a double tax agreement applies which allows someone to work in New Zealand for up to 183 days, and there would be no obligation to withhold PAYE.  

In addition to PAYE an overseas employer may also find itself with FBT and employer superannuation contribution tax issues if it decides to either voluntarily enter into the PAYE regime, or it has a sufficient presence that deems it to be an employer for PAYE purposes.

The complexities of overseas income

Moving on, it's getting towards the mad rush for filing the March 2021 tax returns before the final date for tax agents on 31st March. The more complicated income tax returns will often involve overseas income, and Inland Revenue has just issued new guidance in the form of an Interpretation Statement on how to calculate the foreign tax credits that may be involved.

Now, this Interpretation Statement is very helpful because this is a surprisingly complicated topic. There's a lot of detail involved in this. Firstly, we have to determine is the tax paid, for example, of substantially the same nature as the income tax that we charge. That's not always the case.  Because if the foreign tax is not covered by a double tax agreement and is not of the same nature as income tax imposed under our Act, no credit would be available.

That is something I've seen a little bit more of in relation to charges on pension withdrawals made by the UK and Irish governments. They've been imposing these charges recently where people have made early withdrawals from pension schemes.  The amount imposed can be quite substantial - up to 55% in the case of the UK.  The way the charges have been drafted they are outside the double tax agreements we have with the UK and Ireland. And in both cases, that means that there's no credit for the withdrawal charge. So, a person may face a large charge from either Ireland or the UK and a tax charge here and have no relief for the charges imposed by the Irish and or UK governments. It’s a complicated topic and something to watch out for.

The first booby trap you have to watch carefully where a double tax agreement is involved is what the limits are and whether New Zealand actually has any taxing rights in relation to that income. That's not always the case. But even when you get past those two positions in terms of calculating the credit, you then have to break it down into segments. They must be divided by country and then by type.

If you've had tax deducted of 20% from US sourced income and 10% from UK sourced income, you cannot just aggregate those two amounts and offset them against a single amount of overseas income that you report on your return. You have to actually break it down further than that by country and by type of income. And if you’ve got an attributing interest in a foreign investment fund income, that's a separate calculation as well.

These are surprisingly involved calculations which although people think of as reasonably commonplace, have traps for the unwary in them. So I recommend having a thorough read of this new interpretation statement. It's about 40 pages, and it can be found in the latest Tax Information Bulletin.

The proposed Social Insurance scheme may reduce fraud?

And finally, last week I was talking about the Government's proposed social insurance. This proposes some form of unemployment insurance will be provided, as well as for sickness and other illnesses for employees and contractors alike. As expected, it’s generated a fair amount of interest together with some pushback about costs.

This week an article from Dr Eric Crampton, the chief economist with the New Zealand Initiative caught my eye. He suggested that the new scheme would generate a whole host of rorts. He based his commentary partly on his experience of what went on when a similar scheme was operating in Canada about 30 years ago. Firstly, he suggested employers would put seasonal contractors onto permanent contracts before making them redundant before the end of the picking season. And that would be a win for the employee because they would now qualify for up to six months support.  Alternatively, an employer could sack a person who about to take paternity leave. By sacking them they'd get more than the current paternity parental leave payments of $621.76 per week.

Two things stand out. There isn't much in this for the employer because it is basically fraud. I'm not sure many employers would want to be actively engaged in that. And for that matter, neither would employees. Some of the anecdotes that Dr. Crampton suggested sounded a little bit like, “Well, my friend on Facebook's third cousin said this.”

But he does make a key point. Fraud is a potential issue for the scheme, and I see two areas where it could be a problem. Firstly, the obvious one - trying to make claims when a person is not eligible. The second area is where a person has a valid entitlement, but the amounts claimed are fraudulent because the numbers have been manipulated to over-state the person's income.

Under the proposal, the Accident Compensation Corporation is to be responsible for the running of the scheme.  Now it has a record of managing ACC fraud. Although interestingly, its latest annual report didn't cite any numbers as to how much fraud was detected.

But I think if you are concerned about potential for fraud, then you should draw a lot of comfort from Inland Revenue’s enhanced capabilities following the completion of its Business Transformation programme, because the new START system gives Inland Revenue far more capability to analyse data.

And another thing here, which would be different from the stories of how similar schemes may have operated in the past in other jurisdictions, is that we have now Payday filing. This means there is near real time data of salary payments flowing through to Inland Revenue. So again, attempts to manipulate numbers should be picked up very quickly.

That still leaves the self-employed, where one report suggests under-reporting of income might be as much as 20% of income. But one thing that's going on in relation to the self-employed and especially labour-only contractors, is payments to such persons are increasingly subject to pay as you earn and go through the Payday filing system. So again, Inland Revenue is monitoring payments which gives me some comfort in the matter.

But perhaps counter-intuitively, the introduction of social insurance may mean that we see a reduction in tax evasion. And that is with the opportunity that the social insurance payments represent, people will always want to make sure that they can maximise their benefits. It may well be that persons who previously underreported their income realise that although that might reduce a tax bill it's no good if you fall sick or your contract is terminated. So, maximising their income is in their best interests.

In answer to Dr Crampton’s concerns about fraud, yes, it could happen. But I think the reality is Inland Revenue are far more sophisticated and have the tools to manage this issue than he gives them credit for. And secondly - as an economist, I'm sure he will appreciate this - perversely, there may be an incentive for people to start reporting their correct income to ensure that they maximise the benefit in case they need to make a claim.

Well, that's it for this week. I’m Terry Baucher and you can find this podcast on my website or wherever you get your podcasts. Thank you for listening and send me your feedback and tell your friends and clients. Until next time, kia pai te wiki, have a great week.

Terry Baucher is an Auckland-based tax specialist with 25 years experience. He works with individuals and entities who have complex tax issues. Prior to starting his own business, he spent six years with one of the "Big Four' accountancy firms including a period advising Australian businesses how to do business in New Zealand. You can contact him here.

We welcome your comments below. If you are not already registered, please register to comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.


"it is basically fraud. I'm not sure many employers would want to be actively engaged in that."

The context Eric Crampton gave was seasonal employment.  The track record of RSE employers is far from spotless. 

So it is no stretch to posit that the combination of employer advantage, which is to make certain the cost of an early redundancy, and ex-employee advantage, which is to tide them through to the next season  in taxpayer-provided comfort,  is more than sufficient to come to a handshake- verbal, undocumented and unwitnessed - agreement.  Try proving 'fraud' with That.... 


..exactly my thoughts as I was reading the article.

Historically, much of NZ compliance  (be it tax, welfare etc) relies on the honesty and integrity of the citizen.  This has changed, covid payments being a grand example. The system struggles at present (SFO now bogged down in covid, so other stuff takes a back seat). The danger is that the policing of hand-outs becomes tokenism and the frauds become endemic and unmanageable across all sectors. 


One substantial difference between current ACC and the proposed one is the lack of experience rating. Employers do not want to rort ACC because their premiums will go up. Faking accidents is stupid not only because it's easy to be caught but also because you're then paying higher premiums. 

But who's to say when a redundancy is 'fake'? Nobody involved in the transaction is other than a willing participant.