Dodginess on tax regressions

The Guardian is reporting that Andrew Leigh has a new paper looking at whether firms that pay more tax employ more workers.

In a paper published in the Economic Analysis and Policy journal, Labor’s shadow assistant treasurer finds that companies paying an effective tax rate of less than 25% are shedding jobs while those paying 25% or more are growing their workforce at an annual rate of 2%.

Leigh has used the finding to reject the Turnbull government’s contention that lowering the company tax rate for companies earning more than $50m from 30% to 25% by 2026-27 will boost job creation.

So being a curious type I went and read the paper.

So a few little things occured to me that should also have occured to the referees of the paper and the journal editor but apparently did not.

Andrew Leigh collects data from IBIS World and the ATO’s Corporate Tax Transparency report. As far as I can work out that means he has three years worth of data for about 2000 entities liable for company tax. He doesn’t quite say so in the paper but I think he has used data for 2013/14, 2014/15, and 2015/6. He then averages the data over those years (why?) and then discards all those firms that made a loss – i.e. he truncates the data set (why?). He then estimates a regression where average annual employment growth (with three years of data he has two growth figures that are now averaged) is the dependent variable and effective tax rates is the independent variable. For robustness he adds in Log(Revenue) for a size effect (but didn’t he also use number of employees as a weight to control for size?). This is a mis-specified model for a start – there is a non-linear relationship between size and effective tax rates (ungated version here).

The biggest problem though is that he doesn’t seem to have controlled for profitability. It makes perfect sense that firms that are more profitable would both pay more tax and employ more workers. There is a correlation there. So his regressions should have included some or other lag structure between profit growth in the previous year (or investment) and current employment growth. That analysis is simply lacking. His averaging of the data specifically washes away any effect that we might expect to see. Look at his table 4.

First point: His analysis only provides a result for one of his effective tax rate proxies.

Second point: His adjusted R^2 are very low (except for one regression that I suspect is a typo).

Third point: He starts off with over 2000 observations and is down to 594 and 690 in those regressions with significant results.

Fourth point: He claims to include unprofitable firms in the analysis in equation 6 – but compare the sample size difference from equation 4. There are only 96 unprofitable firms? Compare the R^2 of those two models – down dramatically. So too the size of the coefficient – down from 0.253 to 0.188. I can’t tell if that is statistically significantly different, but it is a big drop and makes me wonder what would happen if he had controlled for lagged profitability?

Fifth point: He cites Tran 1997 on the Book-Tax Income Gap  but makes no effort to control for factors that are likely to impact that gap.

So all up a highly incomplete and inadequate effort that would see an undergraduate econometrics student getting a very average mark were this a mid-term assignment. At the very least this analysis should be performed over a longer time period and probably over an entire business cycle. It is very likely that some firms were still carrying forward losses.

As always – this is not an invitation to slag my good friend Andrew in the comments.

Update: Another interesting observation – the paper was accepted for publication the same day it was received, and went online just 4 days later.

Gives a whole new meaning to speedy turnaround times. A tad suspicious – especially since the journal hasn’t updated the affiliations of its editorial board in several years.

This entry was posted in Economics on the left, Taxation. Bookmark the permalink.

38 Responses to Dodginess on tax regressions

  1. H B Bear

    At least this isn’t as embarrassing as his last effort where the banks were taking over corporate Australia. And Dr Leigh is one of the Liar’s deep thinkers. Truly a student of the original deep economic thinkers, The Legover Man and Wayne Goosesteen.

    Be very afraid.

  2. I think Anglicare made a bunch of wild statements in their so-called review of company tax cuts as well, claiming that the wealthiest 20% are so much better off than those on government benefits, what with all of their tax breaks the former receive: http://www.abc.net.au/news/2018-03-26/tax-concessions-put-welfare-costs-in-perspective-anglicare/9585930.

  3. Tintarella di Luna

    I think Anglicare made a bunch of wild statements in their so-called review of company tax cuts as well, claiming that the wealthiest 20% are so much better off than those on government benefits, what with all of their tax breaks the former receive: http://www.abc.net.au/news/2018-03-26/tax-concessions-put-welfare-costs-in-perspective-anglicare/9585930.

    Anglicare ? you mean the Anglicare that does not permit the sale of raffle tickets because it’s gambling but loses $120 million in the GFC washup because it played the stockmarket? That Anglicare? I think maybe stick to providing the care that changes lives and leave economic analysis to others.

  4. RobK

    Establishing a correlation is only a preliminary step to hint at a mechanism. To base policy purely on correlation is not only fraught with ignorance but has the added hazard of delusions of understanding what is going on.
    So, yes, there is a lot wrong with the analysis as you say Sinc, but also there is no basis provided as a mechanism that would justify any policy. Almost a total waste of time other than to buttress the null argument.

  5. Anglicare ? you mean the Anglicare that does not permit the sale of raffle tickets because it’s gambling but loses $120 million in the GFC washup because it played the stockmarket? That Anglicare? I think maybe stick to providing the care that changes lives and leave economic analysis to others.

    They should, but it’s that moral high ground they occupy that keeps them going.

  6. yackman

    Surely the R-squared cannot be correct? They are so low as to be worthless. Is the figure quoted the statistical significance of the R-squared perhaps?

  7. Roger W

    But when did politics have anything to do with accuracy, truth, careful analysis or statistical responsibility?
    Their ABC and Fairfax, together with The Guardian etc won’t want anything except an anti Coalition headline they can use. Sinc’s analysis won’t even register. And Turnbull has no idea how to fight back anyway.
    Same way all the use of social media etc by Obama and later Clinton is forgotten (and was even praised at the time as an indication of cutting edge 21st century methods) but Cambridge Analytica is front page for weeks.
    Also think constant predictions of the death of the Great Barrier Reef that have been repeated endlessly since the mid 1970’s at least but a James Cook professor loses his job for pointing out the lies.

  8. thefrolickingmole

    The whole thing is a political position looking for some support.
    At its core the “story” is “Companies which are profitable pay more tax and employ more people” hardly a novel finding.

  9. Tator

    Bemused,
    first thing I saw that discredits the Anglicare report is the fact that it was done by Per Capita, the Kouks think tank. INSTANT DISQUALIFICATION for any form of legitimacy in my eyes.

  10. Wozzup

    Let’s see. More profitable companies pay more tax. More profitable companies employ more people. And according to the Wally who wrote the paper the conclusion to be drawn from this is that paying more taxes is good for employment????? Has anyone not told him that correlation is not causation?
    Instead of him asking “Do companies that pay more taxes employ more people?” should he not instead have considered asking :”Do companies that employ more people, pay more taxes?” His was a classic case of putting the fiscal cart before the free enterprise horse. If I were an undergraduate student again I should be seriously embarrassed.

  11. jupes

    Don’t worry Sinc, Morro will be straight onto this.

  12. Megan

    Don’t worry Sinc, Morro will be straight onto thIs

    Just like he jumped all over Shorten’s ridiculous new Imputation Dividends tax.

  13. Louis Hissink

    Submitted and accepted for publication the same day? And peer review????

    Truly an economic miracle.

    Which makes it all a bit sus……

  14. Entropy

    I am regularly astonished at hw often in government you come across people who somehow think prof is a dirty word.
    Yet a company makes a profit, employs more people and yes pays taxes.
    The order is 1. A whole lot of risk and blood and capital setting up a company 2. profit (if lucky) 3. employment, 4.taxes.
    It most certainly isn’t 1. taxes, 2. employment, 3. Profit (assuredly), 4. Steal capital

    What an idiot. And how shameful of the Economic Society of Queensland.

  15. Entropy

    I don’t think these society pubs have too much in the way of quality control. They are desperate enough for articles they take any sort of shit.

    Submit and it’s published.

    Reminds me of a story about a chap who was in the ACT branch of the young liberals. Want a policy? Write one! It would get accepted because none of the other ladder climbers gave a shit about the details.

  16. max

    House Republicans Objected to a 1,100-Page Bill With No Time to Read It . . . in 2009
    Gary North – March 24, 2018
    Printer-Friendly Format

    This is from 2009. Listen to what Congressman John Boehner promised. The Republicans were the minority party in 2009. The promise: “We will post the bills 72 hours before we vote on them.”

    Last night at 7:47 PM, the Republican-controlled House of Representatives posted online the 2,232-page budget bill. Spending: $1.3 trillion. It voted to pass it this afternoon. The story is here.

    When you think “Congress,” think “cons.”

    https://www.garynorth.com/public/17880.cfm

  17. Skin in the game skin in the game skin in the game.

    Tax everything, including the allowances politicians get for their offices, unemployment benefits etc.
    Increase benefits, then tax them by that same amount (say roughly, add 33% then tax 30%)
    Every recipient of government payments should have a taxed amount showing on their statements. That way, when a tax reduction is proposed, EVERYBODY gets to benefit.

    That’ll end the push-back against reducing taxes. In fact, we could become the lowest taxed nation on the planet.
    The numbers NOT paying tax are too high. That’s the problem. Lack of knowledge about taxation is irrelevant. Educating the public about taxation is a total waste of time when more than 50% don’t pay any taxes. They don’t have skin in the game, therefore they don’t care if others pay more. In fact they benefit if others pay more.
    Incentives are all arse about. Take Paul Keatings advice “When push comes to shove, always back self-interest.”
    Make it so that it’s in EVERYONES SELF-INTEREST to reduce taxes.

    TAX EVERYTHING. Pensions, disability payments unemployment payments EVERYTHING.

  18. manalive

    Using data for around 1000 profitable Australian firms, I explore the relationship between effective tax rates and job creation. On average, I find that the relationship is positive — meaning that firms which pay less tax tend to create fewer jobs …

    Reverse causality, Statistics How To (and how not to):
    “Reverse causality means that X and Y are associated, but not in the way you would expect. Instead of X causing a change in Y, it is really the other way around: Y is causing changes in X …”.

  19. benaud

    Sorry dumb question.

    In my limited experience, but correlates with my MBA.

    Companies that are expanding and therefore hiring generally are making a loss on the expanding business segments.

    As it’s expensive to develop plant, process and market share. It’s only as the market matures one expects to make money.

    By removing the lose making businesses he is removing expanding and developing industries. Instead relying on profitable companies in mature markets which could be consolidating or seizing market share.

    Or not, I can be dumb.

  20. Chris

    ALternatively and thinking of the resource boom, a lot of profit or operating surplus means the company can run projects to do the job better or drill out the next deposit in the development program – so there is a ‘discretionary’ burst in employment following a rise in free cash. Might be oeprating or might be capital in allocation, but a rise in free cash enables a rise in employment and also enables a rise in tax.

  21. RobK

    I think both benaud and Chris are right. There are many reasons why a company may have certain employment and profit/tax figures in a given period of a few years. For starters different companies in different sectors will have different labour and capital components anyway, I’d have thought (compare say hospitality vs agricultural). Then there’s the cycles of events were companies will endeavour to finance development in anticipation of an upswing…so no profit but high relitive employment low tax. This is typical for mining..

  22. MPH

    Does anyone follow the scientific method anymore? Or is all modern research just formalised confirmation bias.

  23. Pete of Perth

    That’s faster than getting a climate you are all guilty except academics paper published.

  24. Boambee John

    Baa Humbug
    #2670602, posted on March 26, 2018 at 2:33 pm
    Skin in the game skin in the game skin in the game.

    Indeed.

    GST should be separately identified on every invoice or receipt. But a slab? Make the GST visible on the receipt. In some US states, the sales tax is not on the pricd tag, it becomes very visible when it suddenly is added at the point of sale.

    Tax everything, but refund when appropriate. End PAYE. If every worker had to write a cheque to the ATO every quarter, suddenly income tax would become very real.

  25. Mullumhillbilly

    Significant (p>.01) but meaningless (R^2 less than 0.1). Which part will the ABC ignore for their headline ?

  26. Craig Mc

    Businesses that make more profit pay more tax and employ more people, therefore increasing taxes on businesses will increase profits and increase employment.

    Why, it’s almost Swannian!

  27. Biota

    yackman
    #2670530, posted on March 26, 2018 at 1:08 pm

    They do indeed look more like p-values.

    The journal sounds like one I received an editorial board invitation today based on my ‘immense contribution to the field of biotechnology”. A field I don’t work in and have never published in.

  28. Harlequin Decline

    Using only 3 years worth of data seems very ambitious if they are looking for conclusions that will stand up to scrutiny. It only gives them 2 data points per company given the expected lag between profit and employment.

    The R^2 in the table is so small that any meaningful conclusions drawn can’t really be justified.

    Clearly he should be comparing companies in similar sectors with similar capital/labour mixes and established companies at that to rule out losses carried forward and start up conditions.

    It seems to me it would be better to normalize profitability as taxable profit per employee and tax paid per employee or similar with multiple years of data of companies in the same sector then looking to see if the normalized number of employees(new employees/total employees) increases with increasing tax paid per employee.

  29. Harlequin Decline

    Tax everything, but refund when appropriate. End PAYE. If every worker had to write a cheque to the ATO every quarter, suddenly income tax would become very real.

    Dead right-at the cost of extra bums on seats in the tax department it would radically change the average punter’s outlook on politics.

    They might even start asking the politicians meaningful questions about Government expenditure, possible alternatives, the cost of social programs and start voting in financial conservatives.

  30. 2dogs

    His adjusted R^2 are very low

    Expect Simpson’s paradox. That’s the telltale marker.

  31. Gavin R Putland

    As a Georgist, I don’t approve of taxes on what Adam Smith called the profits of stock. As a Georgist, I recognize capitalists as producers.

    Nevertheless, I am sick and tired of being told that if we want to encourage X, we need to cut taxes on Y, where Y just happens to be important to those who are preaching the message.

    News flash: The surest way to encourage X is to cut taxes on X! E.g., the surest way to encourage employment is to cut taxes on employment!

    (And as a Georgist, I allow myself to be pleased if capitalists benefit from a tax cut on a key input: labour.)

  32. Aynsley Kellow

    ‘Submitted and accepted for publication the same day? And peer review????’
    Even in the annals of climate science, the fastest I have seen (Ben Santer et al attempting to rebut existence of The Pause in time to influence Trump’s decision on Paris) was a month!

  33. None

    This sounds like one of those dodgy papers you lodge at some dodgy outfit that doesn’t actually undergo any real peer review that is simply there for people to pretend that they have published something. One should ask too, if he paid a fee to get it published.

  34. Paul Farmer

    Some good comments here …. I would add that effective tax rates paid by corporations tend to also be heavily effected by tax strategies of accountants and therefore a large part of the noise can simply be the results of how those strategies impact on taxable profit relative to accounting profit which is what he is trying to infer from . A longitudinal study through time on firms within the same industry, hence would likely have similiar tax strategies but that is also a maybe.l, adjusting for tax treatments is the only way to take out this noise .

    The other issue here is pyschology . Essentially Leigh is inferring that a reduction in effective rates would have the same impact upon an employer as a general cut in the company tax rate . There is little evidence for that. How accountants may tinker with effective tax rates and claiming tax benefits through time will have much less dramatic effect if any on how the underlying business operates whereas a significant cut in headline rate may cause investor optimism and general improvement in economy wide confidence . Recent US experience shows this to be something tangible . Exploiting tax advantages at margin and structuring ones tax affairs to minimise it doesn’t have the same psychological effect on boosting confidence . Never will …….. so he is comparing an apple to an orange in the way a company tax cut would work compared to regressing on tax effective rates .

    I would have thought Leigh would have been a Keynesian but seems economists of the left today seem to think the government drives economic activity in all situations of the economic cycle more than the private sector , something Keynes never would have agreed to .

  35. Tezza

    Too clever, Sinc. How about this: The implication of Andrew’s paper is clear: since companies paying low tax rates have weak employment performance and companies paying high tax rates have strong employment growth, we should raise company tax rates (regardless of the opposite trend in other countries), because that would improve employment growth.
    What else could Andrew possibly mean?
    Where is Costello when you need him?

  36. RobK

     Exploiting tax advantages at margin and structuring ones tax affairs to minimise it doesn’t have the same psychological effect on boosting confidence . 
    Over all, I think tax driven management decisions have a very high hidden cost not appreciated sufficiently by the legislators. The cost to the economy of squeezing a bit extra out in government revenue is counter productive, especially if minimization involves convoluted steps. It’s a nightmare. Keep tax simple and small for best returns and employment.

  37. Rococo Liberal

    If the company tax rate was lowered, the survey would still say the same thing. It is not evidence of what Leigh claims for it. There is no evidence that a 30% tax rate of itself makes companies perform better.
    The man is a solid gold wanker.

  38. Tax everything, but refund when appropriate. End PAYE. If every worker had to write a cheque to the ATO every quarter, suddenly income tax would become very real.

    I believe in Hong Kong you had to queue up at a Govt. shop front, in person to pay your entire tax bill with a bank cheque.

    You really have to wonder why Mr Leigh would want to carry on The Goose’s legacy.

Comments are closed.