A tax free threshold?
The support for a tax free threshold in New Zealand appears wider than for almost any other policy. The right supports it, the left supports it – then why is it not government policy, and why do I not support it?
A tax free threshold at $9,000 would cost approximately $3.5bn (according to Patrick Nolan’s NZIER tax cut calculator from November 07), and would give everyone earning over $9,000 the same chunk of money (assuming that other tax thresholds remain unchanged).
Lets discuss why the policy may be popular along the political spectrum:
Now the fact that this effectively hands back the same amount of money to everyone earning over $9,000 (depending on tax incidence of course), or 80% of the population over the age of 15, is what excites people on the left. After all, the tax cut is equivalent to a flat benefit payment to this part of the population – implying that it will reduce inequality. In fact, its inequality reduction impact is obvious when we notice that it will reduce the “average tax rate” for those on low incomes by more than it will for those on high incomes – implying that the proportion of the total tax take paid by the wealthy will now be higher.
The reason for the right supporting such a tax change is significantly different. David Farrar succinctly establishes this point in the following statement about Micheal Cullen refusing to introduce such a change:
This is a pity because I think there is a strong case that there should be no tax on your income until you are earning more than the minimum amount you need to live on. The welfare system then tops people up to that minimum standard, but it is significantly wasteful to tax people just to then give them that money back as welfare
Effectively, the right views a tax free threshold as a method that can be integrated with the benefit system in order to reduce “churn” – which is when the government taxes money and gives it back, which in turn has an administration cost. In this case benefits are provided in order to allow people to achieve a certain level of income. A zero tax rate with a percentage reduction in benefits will allow the government to pay benefits to people who are working without this churn.
Given this the question is – is this policy worth $3.5bn?
Where is the problem?
We know that this policy constitutes a $3.5bn shift of the claim on resources from the public to the private sector. We also know that this policy will reduce inequality and that it will reduce the “churn” associated with taxing people who are still earning benefits. However, we don’t know if such a tax cut will actually be the best use of this $3.5bn.
Tax cuts have another essential role on the “supply side” of the economy. Fundamentally, changes in tax rates can change “supply of labour” depending on the marginal tax rate that these individuals face.
This impact is strongest on secondary earners. Now having a zero income tax at low levels will see more secondary earners enter the labour market – however, if we cut taxes further up the income chain we would see a greater increase in labour supply (especially given our high level of labour market participation!).
Why is this? Well the extra “costs” of being in a job increase in the number of hours you work in that job. Part-time and secondary earners are especially responsive to changes in these costs and the benefit associated with working. As these worker types are becoming increasingly prevalent in NZ (especially on the service side of the economy) a cut further up the tax scale will illicit a greater response in terms of the number of hours these workers will work, by increasing the marginal benefit of this type of work in the face of rising marginal costs.
Another issue is savings. The “marginal tax rate” that an individual faces is important for determining their savings decisions, as income taxes are also placed on the interest returns from savings. The further we place the tax cut up the income scale, the greater the savings response will be.
In terms of the “churn” argument I think an important issue has been missed – the administration cost of introducing another tax bracket. Furthermore. the government separates taxes and benefits as much as they can in order to reduce the administration costs associated with having to go through individual income in any great detail. As a result of these issues and the amount of automation in this process in modern times I’m not convinced that churn is a big issue.
Finally, is it fair? People who work only part time are being given the same amount of money as those who work 50-60 hour days. By moving the tax burden further onto those on medium and high incomes are we really achieving our equity goals?
So why am I against it rather than indifferent
Well, the labour supply argument and the savings argument are important issues on the supply side of the economy. As the benefit system is the best way to achieve our equity goals, changes to the tax system should be made with efficiency goals in mind.
As this tax cut delivers little bang for its buck on the supply side of the economy, I have to stand against it. However, I’m happy to be convinced otherwise 🙂
Minor objection: a good deal less than 80% of the population earns more than $9,500 from taxable sources (I’m pretty sure than benefits are still unaffected by the tax schedule). In fact, the really poor don’t tend to have much in the way of earned income. The 2007 HES has crappy tables and I can’t be bother using table builder, but in the 2004 HES, even decile 4 individuals only had an average wage + self-employed income of $4,700 pa. So this is poorly targeted policy if the goal is poverty reduction – the left are just wrong on this one, I tend to think they only support it because it is a compromise with public demand for tax cuts.
“I’m pretty sure than benefits are still unaffected by the tax schedule”
True – I agree that this reduces the impact in terms of income distribution.
However, luckily this reduces the benefit of a policy I didn’t like anyway – so I’m glad you raised that point.
“So this is poorly targeted policy if the goal is poverty reduction”
It does reduce average tax rates to zero for all those with an income below $9,000 – surely this implies that inequality amongst earned income earners is now lower.
The issue then is – how does this work in relation to the benefit system.
Matt, So you are saying it’s better to lower the marginal rate at higher incomes thus encouraging people to work more. In which case you would be against WFF as high marginal rates there also tends to “trap” at a fixed level of income. I think Gareth Morgan’s Guaranteed Minimum Income idea has merit.
“I think Gareth Morgan’s Guaranteed Minimum Income idea has merit”
The guaranteed minimum income idea is a lot older than Gareth Morgan – he is just the main proponent of it in New Zealand. (http://en.wikipedia.org/wiki/Guaranteed_minimum_income).
In theory the scheme is beautiful. It introduces a flat tax system, but ensures that everyone makes a minimum level of income. We can then add additional components to it in order to achieve other equity goals – thereby tying the benefit and tax systems more closely together.
However, this presupposes that the only purpose of a “benefit” (which is essentially what the minimum income is) is to ensure that people have a minimum living standard. Another view of the benefit is that it provides a security net for the job market.
By paying people a sufficient enough minimum income, we may have a reduction in labour force participation. As having individuals in the labour force increases human capital, there are substantial costs from this reduction in labour force participation.
Given that a reduction in labour force participation is a negative aggregate supply shock, and given that benefit numbers would implicitly increase, this would be a expensive scheme – likely more expensive than its proponents have stated.
Don’t get me wrong – I’m not heavily against the idea. I just think that the tax, benefit, and education systems need to be more heavily integrated – as the impact on human capital of certain scheme is usually under-valued by economists.
My opinion (where I also assume that the purpose of the benefit is to give people a minimum income) is here:
http://tvhe.wordpress.com/2008/02/01/interest-free-student-loans-and-compulsory-schooling-is-there-a-better-way/
Although a tax-free threshold would reduce inequality for people with earned income it would increase it for people without earned income, the latter group is where poverty is concentrated. And very little of that $3bn cost would go to people in poverty, hence my description as “poorly targeted”.
Interesting empirical evidence on the negative income tax (which is the G.M.I by another name) in the US, I think this supports Matt’s point re participation.
Generally, isn’t the main reason for supporting broad-based flat rate tax systems is that you get the lowest marginal rates over the widest range? Whereas a tax-free threshold has very little impact on anyone’s marginal rate – if you’re working you’re probably earning more than $9,500 (14 hours a week at the minimum wage). Admittedly the average rate is important too, it’s possible that if part-time, low-income workers had a really elastic labour supply the tax-free threshold might be an efficient policy in some cases in terms of raising participation.
I think this is a crazy idea – and I have no idea why any self-respecting “right winger” would support it.
This adds another tax bracket, makes the tax system much less flat, much more “progressive”, increase *marginal* tax rates above the threshold, and rewards low-income people for small amounts of low-value work.
If we want to give $3.4 billion back (or however much) by far the most effective measure would be to reduce the corporate tax rate by 3%.
The next most effective measure would be to *cap* individual taxation
– say offering a 100% rebate for every dollar of tax paid over $50,000 for an individual. That is, the marginal rate for income of $150,000 would be 0%.
This would have a number of highly beneficial macroeconomic effects: increasing savings and investment, encouraging high-value participants to return to NZ (or not to leave in the first place), and has a positive outcome for fiscal drag (assuming that threshold is *not* indexed).
hiya Matt and CPW,
Very quickly – last I was on a benefit (quite some time ago) it seemed to be taxed too. This may have changed, or my recollection may have been wrong, but if it is still the case then the TFT would be of benefit to beneficiaries. Agreed that it would only be a small one though.
My personal problem with it (being a leftie and all) is that it gives money to people who don’t need it. My own suggestion was to implement the TFT but to accompany it with an increased top marginal tax rate.
Benefits are taxed, but I’m 90% certain that legislation mandates the net of tax level of the benefit, so if taxes were lowered, the gross benefit would also be lowered, and the net of tax benefit would remain the same. I think superannuation is different though.
I may be simplifying here, but would think that a tex free threshold would simplify the IRD’s work around low earners and minors falling beneath the tax threshold, who would not need to file or be reported. I have to file my daughter’s details with the IRD, just to get her off the 39% tax rate – seems like extra work that would be saved. A tax free threshold would encourage people to think about coming back to part-time work to supplement incomes, pensions etc – maybe not appropriate at the moment, but a good move in stable economic times.
“I may be simplifying here, but would think that a tex free threshold would simplify the IRD’s work around low earners and minors falling beneath the tax threshold, who would not need to file or be reported.”
I’m not sure why a tax free threshold would involve lower compliance costs than a low tax bracket. Ultimately, IRD has to try and find out if people are within that bracket to start with, and firms need to adjust payments based on a lower bracket.
Having less brackets would reduce compliance costs – but I’m not sure a tax free one is any different to a low tax one for compliance costs.