Occasionally, we meet a company that has capitalised their R&D for tax purposes. Most of the time, the reason why that’s been done is a mystery. And yet, surprisingly many accountants advise their clients to capitalise their development costs.
We’ve covered the impact of capitalisation before, so let’s not get back into that, but what that article didn’t really cover is why it might make sense (or not) to capitalise R&D. First, though, a quick definition.
What is capitalisation?
What does it mean to “capitalise your R&D”? Well, the concept comes from something called “capital expenditure”, in accounting terms.
Some expenditure is so-called “revenue expenditure”. If you pay people to work in your business, that is typically a sunk cost. Either you get the benefit out of it right now, or you are paying the wrong people and you’ll never get your money back. Either your secretary does his work, or he doesn’t, but in either case, the salary you pay him is gone - it is not recoverable in any way.
On the other hand, let’s say you buy a big, expensive printer - one of the industrial types that can print a magazine and costs millions of pounds. You expect to be able to use this printer for more than one year, and unlike the cost sunk into an employee, the printer can be sold on to someone else sometime later. For example, if this is an exceptionally good printing machine that costs £10m and is expected to last 10 years, you might assume that next year you could sell it on for £9m, the following year you could sell it on for £8m, and so on. The point of the business then becomes to figure out how to extract at least £1m of profit out of this machine which is “depreciating” by £1m every year.
This machine is what’s know as a “capital asset”, and you’re not allowed to write off the entire cost of purchase of the machine in the year when you purchase it. For tax purposes, you can only write off the “depreciation” amount of the capital asset.
In other words, if you started the year with £15m in your pocket, and then spent £10m to buy the printer, and spend £5m in wages, and generated £15m of revenues, you might say to yourself that you broke even this year - but the taxman would disagree. They’d allow you to count £1m of depreciation for the printer, and £5m of cost sunk in the wages, thus resulting in a net profit for the year of £9m. So you have to pay taxes on £9m of profit even though you didn’t really make any profit - because really your total assets grew by £9m, since you have your £15m back that you started with, and you still have the machine which is worth £9m.
Capitalism is all about accumulating capital (and leveraging it to generate even more capital, etc). If you have the £10m machine (a “capital asset”), you can use it to generate revenue, or you can sell it to someone else to let them generate revenue (and release your capital for use somewhere else).
What about software capitalisation?
So with this in mind, what does it mean to capitalise software development?
Well, the concept there is that when you pay someone to build a piece of software for you, that software, much like a piece of machinery, can be used to generate further revenues. It can also, or so goes the theory, be sold off to someone else. Therefore, the software is a capital asset, and the money that you paid a person to build it is similar to the money you might spend on a machine.
That’s the concept, but what does this mean in practice?
Well, the main effect of capitalisation, as you might recall from the earlier paragraph, is that it makes the current year look more profitable than it would be from a cash flow perspective. So you might have spend £1m developing a new piece of software, but if you capitalise it with a depreciation of £100k per year, for tax purposes you’ve only spent £100k this year. This means that if you spent £1m on developing software that you capitalised, and you generated £200k of revenues from that software in the current year, you’d have to pay tax on £100k of revenue for the current year, even though you’re £800k down the hole!
But maybe I’ll make profit next year
One possible (false) argument for capitalisation would be that next year you’ll perhaps make £200k of profits again, and this will be reduced to £100k by the depreciation on your software next year. However, that is immaterial, since losses from previous years can be offset against profits from the current year.
In other words, in our example, if you didn’t capitalise the software development, you’d have an £800k loss for the current year (no tax payable). If you then made another £200k profit next year, you’d carry forward £200k of that loss to offset it, and so you’d pay no tax for either year.
This is in contrast to the capitalised case where you pay tax on £100k of supposed profit for each year.
And then there’s tax credits
If you didn’t yet see that it makes no sense to capitalise things that don’t need to be capitalised, the final nail in the coffin is the R&D Tax Credits scheme. As covered before, capitalised expenditures do not qualify for R&D Tax Credits. So if you capitalise your software expenditure, you can’t claim R&D enhancement for it - so you lose out on two fronts: you pay more tax, and you don’t get tax credits.
But does it make sense?
If practical matters such as paying tax don’t concern you (lucky you!) you might wish to consider whether software capitalisation makes sense in a purely intellectual sense.
Does it? Well, no, not really. The world has a long history of buying and selling physical things, and we have a good sense of what value they have (though even there, people make dramatic mistakes, such as in market crashes related to physical property like real estate and tulip bulbs). However, when it comes to software, the idea that just because you’ve spent £100k building it, it’s worth £100k, is patently ridiculous.
The world of tech startups is littered with software-building expeditions that cost hundreds of thousands of pounds and delivered nothing of value. Moreover, finding a buyer for a piece of software that’s not generating revenue is a tall order, if not impossible. The revenue-generating ability of software is also extremely variable. In the fast-moving technology world, selling copies of a piece of software today is by no means a guarantee that you’ll continue to do so next year.
On the positive side, software has much more positive upside than machinery. It is possible to spend £100k creating a piece of software that is, based on sales, worth millions every year for years. This is much harder with machinery. Usually, by the time there’s purchasable machinery for it, an industry is already stable enough that such margins are unlikely.
Unlike machinery, software cannot be easily sold to another company in the same industry. Unlike machinery, software does not tend to retain its capital value over time, steadily depreciating each year. Unlike machinery, the value of software cannot easily be calculated. Unlike machinery, software is intangible.
In other words, capitalising software development makes no sense on a practical or intellectual level.
There is one circumstance which we’ve seen where it does make genuine sense to capitalise software development. In some rare cases, software houses deal with customers who have due diligence procedures that include examining their suppliers’ accounts. For example, we know of some banks that will only approve small suppliers who can show steady profits for several years running.
In those cases, a company that intends to make a substantial investment in new software development may choose to capitalise it so as not to risk scaring these types of customers. In other words, capitalisation of software development can help make a company look more profitable than it really is.
Unless that’s your situation, our advice is not to capitalise your R&D.