The revamped R&D tax break: What your start-up needs to know
By Mahesh Sharma
If you are a veritable fountain of ideas, you should be across the Federal Government's R&D Tax Incentive, which returns up to 45 cents for every dollar spent on risky and experimental innovation.
Recently, the scheme was switched to a quarterly basis, with experts predicting that innovators should reap the benefits.
Last year 9,000 companies registered about $20 billion in R&D spending, according to AusIndustry general manager David Wilson, and the government provided about $1.8 billion in relief.
Big companies spent the most on R&D, while smaller companies made up the lion’s share of the numbers.
Wilson expects these numbers to increase as the 2012-13 financial year was the first year the government sweetened the R&D tax incentive for unprofitable small businesses with turnover under $20 million.
But, as is often the case, the devil is in the detail.
RSM Bird Cameron R&D adviser Stephen Carroll says that to access the quarterly credits companies must get a notification number from Innovation Australia.
This number is required to claim the quarterly credit and adds a significant burden to SME requirements.
"A company will still be required to make its annual R&D registration with AusIndustry and go through the existing registration and claim process," Carroll says.
Additionally, first-time applicants can't claim the quarterly credits in their first year, which is a huge disappointment for start-ups.
"The intention is to assist start-up business to access cashflow, but they are specifically excluded from the program," he says.
So is the scheme really suited to start-ups? And how have innovators fared using the tax break up until now?
Achieving the impossible
Adioso co-founder Fenn Bailey is completing the R&D Tax Incentive application for 2012-13, after it successfully claimed a tax refund last year.
He and Tom Howard boast one of the few travel search websites powered by proprietary technology, distinguishing it from rest of the industry whose airfare search sites are powered by Google's ITA software or the decades-old global distribution systems.
This seemingly impossible task – to develop a new technology for the travel industry – perfectly suits the R&D tax incentive reward of experimental and risky products.
"There's no documentation as to whether it's mathematically feasible to do what we're trying to do,” according to Bailey.
“So we hit a lot of dead ends when we're developing this sort of stuff, in terms of developing algorithms that are capable of answering the queries we're trying to answer."
One of those dead ends forced them to scrap an early version of the site, which couldn't scale to handle their ambitious travel search goals – a search such as "Melbourne to Asia" for every flight ever scheduled and every combination of flight routes.
"I haven't done the numbers recently but it's in the realms of billions, if not trillions, of combinations that need to be evaluated to answer that question,” says Bailey.
“And our target is to be able to answer all search queries in under a second, so you need to be able evaluate a trillion combinations effectively in under a second.”
"A computer can't do that naively by just considering all trillion combinations, so there's some interesting algorithmic optimisations that allow that to be done to allow that to be feasibly answered in a reasonable amount of time.”
Experimenting with success
Accountancy firm TCF Services is an R&D tax specialist and lodges tax returns for over 50 software clients – less than half of its overall customer base spanning industries such as manufacturing, and medical science.
On July 1, 2011 the government redefined research and development by introducing a new R&D tax regime.
Previously, systematic investigative and experimental activities (SIE) simply had to be innovative or risky, but not both.
Experimentation is key, which has shifted the focus from projects to activities: not what you do, but how you do it.
It has forced entrepreneurs to re-open their high school textbooks to explain their R&D in the scientific method: hypothesis, experiments and outcome.
Andrew Flick, R&D business services manager at TCF, says: “It's more micro detail. So the actual writing of the software is the project but within the project you have to find an element that is more challenging and difficult, that is experimental, to be able to have a project to base it on.”
“It's a higher bar of R&D now to be able to qualify. But you're getting more money for that qualification, which is not bad.”
“So those who are doing higher level R&D but are a bit more marginal, it's knocked some of them out of the scheme.”
Flick says start start-ups won't qualify if there's no technical innovation. For example, a new business model or marketing strategy, or if an entrepreneur doesn't pay themselves a wage (that is a business is built on sweaty equity).
"We don't have any R&D getting knocked back but that usually happens when someone tries to do it themselves," he says.
"We pick up a lot of clients that tried to do it themselves but ran into a lot of problems."
The proof is in the pudding
PwC tax partner Sandra Mason described the innovation as “an experimental activity whose outcome cannot be known in advance,” citing section 355-20 of the R&D Tax Law Amendments, R&D Act 2011.
It’s crucial to document the R&D at the time of experimentation, which could test user acceptance on some new code, or an engineer or R&D manager overseeing production on a factory line.
It should answer a number of questions:
It’s difficult to claim without documentation, which is important because the ATO vigorously enforces the rules.
"They do scrutinise and have a process called the ‘compliance continuum’, where they have a protocol program that systematically reviews claimants of the incentive program," Mason says.
"There's teams of assessors in each state and federally that review claims."
To self-assess or consult?
Mason stressed that companies should seek external advice. AusIndustry general manager David Wilson said the scheme was simplified for start-ups to self-assess their tax needs and his organisation also offers a number of resources.
Melbourne start-up Rome2Rio is a first-time applicant and aims to multiply the value of a recent $450,000 investment to fund the development of a B2B model for its multi-modal travel search engine.
The core Rome2Rio algorithm powers the "multi-modal" searches, which instructs customers how they can get from A to B using a range of transport options.
Rome2Rio co-founder Michael Cameron said they have developed the algorithm to return search results faster and with greater accuracy
He is confident of qualifying with the help of AusIndustry and their peers – Rome2Rio shares an accountant with Adioso.
"We have a source control system so every bit of code we write is checked-in – we have logs, comments and email records – and the advice we have is that it's a pretty good set of evidence to make your submissions,” Cameron says.
“So I think from that point of view for the next year we're pretty well set up."
Five tips for R&D success
Top 5 Most Read