A robot couldn’t make this up . . . or could it?

12/02/2017 | by Oliver Shah @ The Sunday Times

st

It was the kind of property that changes hands every day: an office block in Newcastle, leased to a firm of regional lawyers for 15 years. Priced at £10m and offering an annual return of more than 6%, it should have been attractive. But one of the prospective bidders, Trinova Real Estate, turned it down straight away.

Trinova was not concerned about Brexit or the local economy but the potential impact of automation on the legal profession. The property company’s bosses had been following the work of Mike Lynch, the controversial tech entrepreneur whose portfolio of investments includes a machine learning start-up used by the magic circle law firm Slaughter and May to process deal paperwork.

Linus Forsberg, one of Trinova’s partners, said they decided the legal profession was “very much under threat from artificial intelligence [AI] and automation”.

He and his colleagues wondered whether the occupier of the Newcastle building would even exist in 15 years.

If their thesis is right, the march of automation could have vast implications for the broader property market, whose staples outside London are office blocks leased to small firms of accountants and solicitors. It would not necessarily mean the death of business parks, which accommodate a range of businesses such as call centres and logistics hubs as well as services companies, but it could hit prices in certain towns and cities.

“We’re talking about a contraction in the investable universe,” said Sam Resouly, another partner at Trinova. “Certain bits of real estate that are currently considered investment grade won’t have a use in the medium-to-long term. There will be a shrinking of the investible market and a significant destruction of value for owners of those properties.

“You may be sitting on 15 years’ income now and feeling fairly good about life, but in 15 years’ time you’ve then got a building that is structurally redundant.”

The legal industry is already experimenting with programs that carry out work formerly assigned to junior lawyers and trainees. Luminance, the system backed by Lynch’s private equity firm, Invoke Capital, claims to be able to “read” hundreds of pages of complex transactional information a minute, identify risks and sort documents.

The law firm Dentons has created a tech incubator, Nextlaw Labs. Its projects include a Brexit search engine that reviews contracts for provisions that could be hit by Britain’s vote to leave the EU. Pinsent Masons has developed a tool called TermFrame, which guides lawyers through deals and points them towards the right legal precedents, and Linklaters has signed a deal with an AI provider, RAVN Systems, for a document analysis program.

Edward Chan, a partner in Linklaters’ banking practice, said: “Some provisions can be read by the machine, some can’t. That’s the thing we’re struggling to understand: you look at the technology today and it doesn’t replace the human — it supports the human, in many respects.

“There are a lot of things it still can’t do. But it can probably do more than most of my colleagues would like to believe.”

According to the accountancy giant Deloitte, tech developments have already contributed to the loss of 31,000 administrative and secretarial jobs in the legal sector, although it said there had been a net increase of 80,000 jobs due to the hiring of more highly-skilled barristers and solicitors.

The consultant warned that another 114,000 jobs were likely to be automated over the next 20 years. That’s a fraction of the total 1m business services jobs that Deloitte believes are at “high risk” of being replaced with robots over the same period. Bookkeepers, payroll managers and wage clerks are most at risk, Deloitte said, with software “likely to cost less than an onshore, or even offshore, staff member”, and also likely to be “more predictable and consistent”.

The phenomenon looks set to affect both white-collar and blue-collar workers. In a report last month, the management consultancy McKinsey said that while fewer than 5% of occupations had activities that could be entirely replaced with automation, 60% had significant elements that could be robotised.

McKinsey put the wage bill that could eventually be automated in the five biggest European economies at $1.7 trillion (£1.4 trillion) — the equivalent of 54m jobs.

The repercussions could be felt far beyond accountancy and law. Carl Frey, co-director of the Oxford Martin programme on technology and employment at Oxford University, has said that the roles of loan officer, credit analyst, property broker, benefits manager and postal clerk are the five paying more than £40,000 that could be most easily automated.

ROBOTS IN NUMBERS

  • 31,000administrative and secretarial jobs lost due to tech, according to Deloitte, which also says . . .
  • 114,000jobs could be automated in the next 20 years
  • 5%of occupations have activities that could be entirely automated
  • 60%of occupations have significant elements that could be roboticised

The optimists’ argument is that automation will spur increases in global GDP — by up to 1.4% a year, McKinsey predicted — and the creation of new types of work. The consultancy cited the newspaper industry as an example, pointing out that printers at The Times staged a revolt when the publication switched to a steam-powered press in 1814. The prototype, developed by German engineer Friedrich Koenig, printed 1,100 pages an hour — five times as many as its mechanical predecessor.

By 1890, the arrival of electricity and new processes allowed the New York Herald to print 90,000 copies an hour of its four-page paper, with colour illustrations. Rather than destroying employment, machinery powered a huge expansion of the industry.

However, investors such as Trinova are concerned that the speed of advances in automation and a drab economic backdrop could crunch commercial property values in the medium term, particularly in regional cities. The property agency JLL has predicted an increasing polarisation between “platinum prime” centres such as London and Manchester, and less desirable commercial areas where demand will become weaker.

CBRE, another agency, analysed the potential impact of automation on the US property market. It found that up to 18% of American office buildings were in danger of becoming redundant due to headcount reductions and company closures.

CBRE’s report said that mature markets in big cities, such as New York and Houston, would be “relatively immune” to the trend because they attract businesses with higher-skilled staff. It described automation as a “non-trivial risk for all markets” that grows as the city size becomes smaller.

Nick Axford, head of research at CBRE, said: “Clearly, there will be some impact on values, and there will be winners and losers, but it’s impossible to speculate on what that means for the market as a whole. To the extent that automation is going to reduce the amount of space that’s needed — and that’s still very much an open question — we’d expected the impact to fall more on secondary markets.”

The office block in Newcastle rejected by Trinova is typical of the kind of stock that fills the portfolios of British pensioners — whether or not they know it.

Institutional fund managers that run their money have increasingly been priced out of prime areas by overseas investors with deep pockets since the 2008 crash. The Newcastle property’s owner agreed a sale to a private investor at £10.8m last month, indicating that there is still decent demand from buyers who do not share Trinova’s fears.

David Paine, head of real estate at Standard Life Investments, said advances in technology had been changing the workplace for “decades”, initially driving the move to out-of-town business parks and more recently the trend for city-centre shared working spaces.

He acknowledged that automation might result in “more efficient use of floor space per head”, but said: “I come at it more positively. Change is constant in the built environment, and those centres that are most susceptible to that change are likely to find other uses.”

Paine predicted obsolete offices would be converted into flats or restaurants, although he said owners in less affluent areas would find this more “challenging” than others.

“I don’t see this as a negative change in the sense of a fundamental change to values,” he said. “It’s a transfer of values from one type of use to another.”

Lee Elliott, head of research at the estate agency Knight Frank, added: “It’s not a foregone conclusion that because a small professional services firm gets impacted over time by automation, secondary office stock becomes obsolete. Offices will become places of social interaction rather than being full of banks of desks.”

For now, the likes of Paine and Elliott are in the strong majority. Owners of swathes of commercial property across Britain must hope they are right.

Comments are subject to our community guidelines, which can be viewed here.

http://www.thetimes.co.uk/article/a-robot-couldnt-make-this-up-or-could-it-v3kxhbzm0