This is part two of tech entrepreneur Matt Barrie’s in-depth look at taxation issues in relation to employee share schemes.
In part one he explained how employee share schemes work and how they are used especially by tech companies and startups as an enticement for hiring potential employees.In part 2, he looks at how Australian authorities have tackled the issue and how it could be implemented in the future.
1995 to 2009: Division 13A
Prior to July 1, 2009, the taxation situation in Australia was very similar to the current US situation. The default taxation position was deferred; in other words, you were assessable on any discount given at the cessation time, but you could choose to make an election (known under the Division 13A rules as a ‘139E election’) broadly similar to the US 83(b) election I outlined earlier.
If you chose to make the election, you included the value of any discount you received between the market value of the security and the price you paid in your income tax return in the year the shares or rights were granted. This discount formed part of your cost base of the shares for when you finally got around to selling the shares or rights. The government also gave you a $1000 concession if you paid tax upfront, if certain conditions were met.
The 139E election allowed you to avoid that nightmare scenario I talked about before where tax had to be paid every time a share or option vests. So before 2009, most ESOP options or shares didn’t attract tax until they were sold. If you held for a year, under the Capital Gains Tax rules, on future disposal the top rate was 23.25% which was half of the top income tax rate of 46.5%.
If you didn’t make a 139E election, and you had a qualifying employee share scheme, then the “cessation time” which was defined to be, for shares, the earlier of (a) the date of you ended up selling the stock; (b) the later of the time when disposal restrictions were lifted and the time when the shares could no longer be forfeited; (c) the time when employment (in respect of the share) ceased; and (d) 10 years. For rights, such as options, it was similar but included the earlier of (a) the disposal of the right (b) when employment (in respect of the right) ceased (c) if the right was exercised and there were restrictions on either the disposal of the share or the share could still be forfeited – when the restriction was lifted or the share no longer subject to forfeiture (d) exercise of the right or (e) 10 years.
The rules weren’t perfect but they were workable for the most part.
Gillard goes after the investment bankers and wipes out the technology industry: Division 83A
In 2009, the Gillard government changed the rules on the taxation of employee share schemes so that:
1) You could no longer elect to be taxed up-front in relation ‘qualifying schemes’ otherwise subject to deferred taxation; and
2) The ‘deferred taxing point’ focused on vesting rather than exercising (that is, employees could no longer control the timing of their income tax liability by holding off on the exercise of their options until they were in a position to cover the resulting liability, for example, in the event of an exit such as a trade sale of the business or IPO).
I’m not sure of the exact reason why Gillard thought this was a good thing, but I presume that this was a response to the private equity boom in the early 2000s that resulted from governments globally setting low interest rates to fuel cheap credit in the wake of the dot com crash at the turn of the century. I think this was probably a misguided attempt to tax senior executives from investment bankers and private equity who had experienced a short-term windfall at the height of the private equity boom.
Of course, by the time this legislation was enacted in July 2009, the markets had turned (as they always do), and the global financial crisis had hit, wiping out a lot of those banker’s profits. However, in the process of enacting this legislation, the Gillard government had wiped out the primary means of remuneration for the fledgling Australian technology industry. This couldn’t have come at a worse time as trying to start a company and attract good talent was a very difficult task coming out of the global financial crisis. This was somewhat ironic for Labor, because the whole purpose of an employee share schemes is to distribute wealth from owners to workers, and by passing this legislation all that happened was companies stopped issuing employee share schemes – in other words, the owners just kept the stock to themselves.
The way in which the changes work, and I’ll focus here only on deferred taxation, was that there was now only two ways you could defer tax on employee share schemes. The first was if you made a salary sacrifice to buy stock, but you could only salary sacrifice at most $5000 worth of shares per year – which is pretty useless for attracting high quality staff. The second way allowed employees to defer tax only if the employee had what they called “a real risk of forfeiture” of their grant. By “real risk of forfeiture” it meant that there had to be real risk that the employee could not receive their grant, such as a performance hurdle or minimum terms of employment (with the ATO’s rule of thumb being at least 12 months in this regard). I’ll come back to this in a second.
The thing that really screwed everything up was that the deferred taxing point in time to when there are no longer any genuine restrictions on the exercise of the right, or the resulting share from the right, and there is no real risk of the employee forfeiting their right or underlying share. So basically, tax is only deferred until vesting – which is the nightmare scenario. Employees have to pay tax well before they can sensibly sell part of the grant to pay for the tax.
In order to qualify for deferred taxation under both the pre and post 1 July 2009 regimes, an employee participant cannot have more than 5% of the company. The problem with this is hiring employees in the very early stage will cost you potentially a lot more than 5%. Likewise, hiring a heavy hitter at the VP or C-level at almost any stage could easily cost you that much. Hiring an experienced CEO into a startup, for example, is usually going to cost at least 10%. The problem with the post 1 July 2009 regime was that tax was automatically deferred, with no chance for the new hire to make a 139E election and pay tax up-front at a low current value.
There were two fundamental misconceptions in bringing in 83A. The first was that tax revenue was being lost under Division 13A, when it was merely being delayed. The second was that it overlooked that employee share schemes (ESS) drive productivity, which increases the taxable base.
These changes effectively ended the use of ESS in Australia. This was a huge problem for startups, and resulted in a significant drag on activity in the start-up sector and acceleration of movement of startups and talent going offshore, to the detriment of the Australian economy. Ironically this not only reduced the amount of tax paid from discounted grants, but also from businesses that could not attract good enough talent to get off the ground, or those that left the country.
The proposed changes in the draft legislation
In 2014, the current Coalition government announced changes to the taxation of ESS. Draft legislation has been released which is intended to be introduced as from July 1, 2015.
The first thing to note here is despite the rhetoric from the government, they have not “rolled back” the changes to the prior position we had in Australia where you could do a 139E election. That would have been preferred to what has been proposed.
ESSs will be established as either a deferred or upfront tax scheme through their governing rules.
An employee who receives a grant under an upfront scheme will be eligible for the current $1000 exemption of the amount of the discount subject to tax, provided that their income is not greater than $180,000 in that year.
The maximum ownership and voting restrictions will be raised from 5% to 10% (note, however, that this 10% is now calculated on a fully exercised basis).
The tables in Division 13A ITAA36 used to value unlisted rights are updated to use a risk free interest rate of 4%, dividend yield of 4% and volatility of 12%.
An employee who is granted rights under a deferred tax scheme will be able to defer until the rights are exercised, or employment ceases, up to a maximum of 15 years. On exercise, income tax will be payable on the difference between the market value of the right and amount paid to acquire it. CGT will be payable when the shares acquired by exercising the right are sold, on the difference between the sale price and the market value of the shares on the exercise date.
If there is no “real risk of forfeiture”, income tax will be payable on the discount in the year the shares are received. There is a $1000 exemption available under certain conditions.
When there is a real risk of forfeiture, the tax may be deferred until that risk ceases to exist. At that point in time, income tax is payable on the discount from the market value paid to acquire the shares. In addition, CGT will be payable when the shares are sold, with the taxable amount being the difference between the sale price and the market value of the shares when the risk of forfeiture ceases.
A “startup” concession is introduced so that an employee does not include a discount on ESS interests acquired in their assessable income if the employee holds the interest for a minimum of three years, and that the company (a) is not listed on a securities exchange (b) has been incorporated less than 10 years ago (and all group companies similarly less than 10 years), (c) the company’s aggregated turnover is less than $50 million and (d) the company is a resident Australian taxpayer. Instead this is treated as capital gains at later date.
If shares are issued with a discount of no more than 15% to market value then that discount is exempt from income tax and the share once acquired is then subject to a taxable capital gain on the difference between the sale price and the market value of the shares at the time of acquisition.
If rights are issued, provided that the exercise price of the right is equal to or greater than market value of the underlying share, then the discount is not subject to upfront taxation and the right, and resulting share once acquired, is then subject to capital gains. So basically if options are given for free (but with a strike at least equal to the underlying market value of the share on the date of grant), the option premium is taxed later under capital gains and not as income in the year of the grant.
First, let’s start with the good.
The proposed changes relax the significant ownership and voting rights limitations by doubling the existing 5% test to 10%. Note however that the goal posts have moved. Under the current regime, you could own 4.99% and then receive a grant of options over an additional 5% and still qualify as it focusses on the actual shares that you hold or are able to control. The fact that you can potentially exercise the options and obtain the additional 5% equity does not mean you breach the 5% threshold. Under the proposed regime, if you hold 5% and you receive options over an additional 10%, then even though you haven’t exercised the options (and may never) you are taken to effectively hold 15% for ESS purposes breaching the 10% rule and you don’t qualify.
Second, in ESS deferred schemes where income tax is deferred, the maximum deferral period has increased from 7 to 15 years, to give people more time. This is good. Due to the anaemic funding environment for Australian startups, bootstrapping is the default path for our companies, and as such it takes longer for them to get to the promised land. You’d be in a terrible situation if you had a grant that was way in the money in a successful Australian company that was still private after seven years and had to figure out how to pay tax while there was no market to sell your stock.
More significantly, the changes proposed in the draft legislation alter one of the taxing points from the point at which a right can be exercised (vesting) to the point in time where the right is actually exercised.
At least this is the stated intention, however the Exposure Draft Memorandum isn’t exactly clear on this issue, and it needs to be clarified to ensure that this intention is actually achieved. For example, on page 10 of the explanatory materials to the bill it states:
1.36 Where an ESS right is subject to deferred taxation, the taxing point occurs at the earliest of one of the following times:
- when the employee ceases the employment in respect of the acquired right;
- seven years after the employee acquired the right;
- when there are no longer any genuine restrictions on the which they acquired the right;
- when there are no longer any genuine restrictions on the disposal of right (for example, being sold), and there is no real risk of the employee forfeiting the right; or exercise of the right, or resulting share being disposed of (such as by sale), and there is no real risk of the employee forfeiting the right or underlying share.
1.37 This Schedule amends the second and fourth of those taxing points so that the taxing point occurs at the earliest of one of the following times:
- when the employee ceases the employment in respect of the acquired right;
- fifteen years after the employee acquired the right;
- when there are no longer any genuine restrictions on the which they acquired the right;
- when the right is exercised and there is no real risk of the disposal of right (for example, being sold), and there is no real risk of the employee forfeiting the right; or employee forfeiting the resulting share and there is no genuine restriction on the disposal of the resulting share (if such risks or restrictions exist, the taxing point is delayed until they lift).
The problem with this draft is that item 3 is not changed – options vest before they can be exercised, so the taxing point according to this wording is still on vesting for rights issues (“when there are no longer any genuine restrictions on the which they acquired the right”) unless you are careful to specifically set up your option plan so that employees cannot sell their options even though they vest, they instead can only exercise their options. This somewhat defeats the purpose of vesting. I hope that this point is clarified because it appears to be contradictory to the stated intention of the proposed changes.
Finally, if you meet the startup rules then share and option grants prima facie are treated well, although to an employee the acquisition of rights are less attractive than shares because deferral of tax on the discount is only available if the exercise price of the rights is at least equal to the market value of ordinary shares, while shares can be issued at a discount to market value of up to 15% can be acquired without triggering any liability to upfront tax.
The exclusions for the startup concession seem designed to exclude every meaningful technology company in Australia.
Excluding companies with a turnover of more than $50 million
There are several things bad with this exclusion.
First, I disagree with the premise that more advanced companies can pay more and therefore should not be eligible for the concessions. The technology industry is globalised, and companies like mine (Freelancer) need to compete for talent with the likes of Google, Uber, Palantir and Facebook. US companies are hoovering up as much talent as they can in Australia to take back to Silicon Valley – talent that can reap the benefit of 83(b) elections. Equity is the primary remuneration scheme and it’s an unfair playing field for Australia’s up and coming technology industry to be penalised for operating in this country.
Second, to base this on aggregated turnover is just sloppy policy. This is prejudicial to some business models like marketplaces, financial services, payments systems and e-commerce retailers of third party products where the turnover is very large but the margins are wafer thin. An online payments business with $50 million in turnover might have only $500,000 in revenue because it operates on a 1% margin. A marketplace with 10% commission might have only $5 million revenue.
This exclusion simply hasn’t been thought through and hurts our best technology companies who are at a critical time in their growth; companies like Atlassian, Envato, Kogan and my company, Freelancer.com.
Excluding listed companies
The restrictions imposed to access the startup concession are bad policy. First, all listed companies, such as my company Freelancer.com, are excluded. The rationale in the explanatory memorandum is “a listing on an approved exchange allowed easier access to capital and allows for the value of a company to be more easily ascertained and that it is in a more advanced period of development where concerns about ESS compliance costs and liquidity are less prohibitive”. This assumption is patently untrue in the Australian environment. As I have repeatedly written before, due to the disaster in the venture capital industry in Australia, the ASX is now the primary financing source for Australian technology companies.
Just like the resources industry did before them, early stage Australian technology companies are heading to the ASX because a viable venture capital industry does not exist. In 2013, $726 million was raised by Australian technology companies on the ASX via primary issuances. In 2014, eight software companies went public on the ASX with $796 million in primary issuances, another five technology companies went public with over $1b in primary issuance, and a further 238 technology companies either used a backdoor listing or raised via secondary issuance over $234m.
Compare this to the Australian venture capital industry which in the whole of 2013 only raised $155m via three funds, and invested a total of $111m in 69 technology companies according to the 2013 AVCAL Yearbook.
To exclude listed companies just because they are listed is absolute madness.
Excluding companies 10 or more years old
Not only does the company need to be incorporated less than 10 years ago, but all companies in the corporate group need to be less than 10 years ago. I can’t see the purpose of this other than to be prejudicial to companies that have not taken in external capital and have instead bootstrapped. This rule excludes great Australian companies like Atlassian, Nitro, Campaign Monitor, and so on. Furthermore, if your startup acquires a company that is more than 10 years old you also get caught out by this rule as no companies in the group can be more than 10 years old.
Risk of forfeiture not defined
The legislation makes reference to the need of a “real risk of forfeiture”, but does not clearly define what this is. Examples are given of performance hurdles or minimum tenure. From consultation with tax professionals, I’ve been told that this is a grey area and a rough rule of thumb for minimum tenure is 12 months. It was unclear from this consultation whether this automatically requires not just a minimum one year term for vesting agreements but whether a minimum one year cliff was also required.
Minimum three year holding period
Similarly, the three year minimum holding period introduces complexities in offering grants to directors of the board, advisory board members and other consultants where typical vesting agreements might be two years with no cliff due to the higher turnover nature of the roles.
Determining market value for illiquid private companies where no market exists
In Part I of this article I explained how not all shares in startups are the same, and that in particular financial investors are almost always issued preferred stock, which subordinates both the rights and financial returns of the ordinary stock. As a result, the value of ordinary stock is often vastly less than preferred. Indeed, there are plenty of precedents where companies have sold where the liquidation preference much greater than the sale price of the company, so that common stockholders received nothing even though the valuation of the company might be north of $100m.
The major issue here is how startups value the market value of ordinary stock in those circumstances when it is highly likely the only points in time where a price has been determined is based off the share price of the Series A or B of the preferred stock. I’ve commonly seen in the US a realistic discount of around 90% applied, which is substantially larger than the discounts put forward in the safe harbour tables. The schedule also introduces a new power for the Taxation Commissioner to approve market methodologies for valuation of rights and shares that worries me.
If you’re employed by a successful Australian company that is still privately held and are sitting on a grant that is way in the money, you’d better be careful not to get fired or otherwise leave your job unless there is a way you can find a buyer for your stock. If that happens, you’ll be in a world of pain because you’ll be immediately faced with a taxation point – and potentially be in a situation where you’ll be unable to sell simply because there is no market for the stock, especially the subordinated ordinary stock. You’re not able to sit on it and pay tax when there’s an exit for the company down the track. This is quite nasty.
If you do not meet the startup rules you continue to face the nightmare vesting scenario that Gillard introduced in 2009 if you are given a share grant. This is because one of the eligible taxable points in time is “when there is no real risk of forfeiture of the shares and any restrictions on the sale are lifted”, i.e. at vesting. So any discount is treated as income tax at each vesting point. For example, if are given the right to buy shares at $1 in a grant over 4 years with a 1 year cliff and monthly vesting after that, then if the stock rises steadily from $1 over time, then you will have 37 separate taxing points where valuations on the stock will need to be conducted (one at the cliff anniversary and then each month for 36 months to the fourth year after that). This is a nightmare.
Likewise if you don’t meet the startup concessions and you are given rights, you will generally be taxed at each vesting point unless the scheme is designed in such a way that the underlying shares are subject to additional restrictions even though you are vested. The relevant taxable point is similarly “when there is no risk of forfeiture of the rights and any restrictions on the sale are lifted”. I am not sure if the Explanatory Materials to the proposed legislation are in error over this point as it seems contradictory to what I believed the intention of the changes is supposed to be which is to provide for exercise as a relevant taxing point but as currently drafted it looks like this is the case.
Additionally, if those rights are provided at a discount (for example, given options for free without paying the premium) then you will be assessed for income tax at the time of the grant on the discount. You won’t get the ability to defer it like eligible startups can.
Even worse, rights acquired under deferred tax schemes by companies that do not meet the startup concession will be subject to income tax, and not to capital gains tax, on exercise. Neither the Division 115 ITAA97 CGT 50% general discount nor the Division 152 ITAA97 15-year 100% exemption from capital gains tax are available. So this is further explanation about why the government isn’t repealing the 2009 changes and is instead introducing this new legislation – they’re making you pay at least twice the tax you would have pre-2009. Now the full amount of the increase in the capital value of the right since acquisition will be subject to income tax. So it’s a horror scenario for people to be issued stock options.
When I started writing this article I didn’t intend it to be nearly 10,000 words long. This goes to show just how horribly complex the Australian government has made taxation law around this issue. I haven’t even started to cover issues as basic, yet critical as valuation, conflicts throughout the tax law on the issue whether it be FBT, buyback rules, Division 7A (deemed dividend rules), general CGT rules, anti avoidance rules and value shifting rules. Different departments even within the ATO differ in their view of such basic issues.
So complicated is the issue that the drafting of the explanatory memorandum for the proposed new changes has mistakes (such as the clarification on taxation on the vesting of rights I explained above). All of this needs to be a lot simpler, in one section.
Fundamentally, the government here can’t see the wood for the trees and is not doing a cost-benefit analysis. Employee share schemes are vital to Australian technology companies being able to attract and retain talent in a highly competitive environment. Done well they create alignment between employers and employees and greatly increase productivity which delivers two great benefits to the country; it grows taxable wealth and distributes that wealth from employers to employees.
The measly $200 million “budget saving” that Labor made in 2009 is more likely a huge net loss if you consider the drag on the technology industry and economy, the lack of competitiveness it’s created and the contribution to brain drain of both talent and technology companies leaving the country. The tax it did collect is also directly a false saving – changing the rules so workers would pay tax on the value of the shares they were issued, even though this would be years before shares were sold and gains realised in most cases.
The government is trying to be tricky by saying they are “reverting the mess that Labor created” while carving out the vast bulk of companies in the technology industry who employ the vast bulk of employees in the industry – for whom getting this legislation fixed really matters to be competitive at the global level.
So the simple answer is no, these proposed changes don’t reverse the damage done in 2009 and don’t fix the issue with the taxation of employee share schemes in Australia, and this will continue to be a drag on the industry and the economy.
Employee share schemes should just be a simple, easily understood and applied tool that is off the shelf and cheap to implement. What we’ve instead created is costly bureaucracy which is a massive distraction from the real task at hand – creating world class technology companies.
With thanks to David Kenney, Partner at Hall Chadwick, for his assistance as a sounding board and in reviewing this article.
This is part two of Matt Barrie’s submission to Treasury in response to the exposure draft legislation: Improvements to the taxation of employee share schemes. Submissions closed on Friday, February 6.