How much is the 'free' internet really worth to us?

On my morning commute last week I decided to dip into a bit of 18th-century philosophy: Hume's fork, Kant's Critique of Pure Reason. Then I veered off into some physics: Newton's law of universal gravitation, Einstein's theory of general relativity. But what's even more remarkable than my geekiness is the fact that I did all this on my smartphone. The articles were on Wikipedia. And I started out not with a planned reading list but with Google.

In those ancient days before smartphones, wifi on public transport, and, of course, the internet, I would have needed to go to a library to read up on such an eclectic range of topics. Or I would have needed to carry a number of books around with me, carefully selected before leaving the house. We all know that digital technologies have changed our lives over the past decade. But what is harder to pin down is how much this benefit is worth to us economically.

Leaving aside for a moment the issue of advertising and our personal data, there is no upfront charge to use a search engine like Google. Wikipedia is, of course, a public resource. But this means there is no price tag for economists to use in order to impute the value we all derive from such new services.

But that doesn't mean economists have given up on estimating it. In 2010 Yan Chen from the University of Michigan conducted an experiment to see how much time people save through searching for information online relative to the older methods. She found that the average online search time for a given task is seven minutes, versus 22 minutes for the offline search. We should perhaps take the research with a pinch of salt given that it was partly funded by a Google research grant. Yet there seems little reason to doubt the basic result that online search is around three times quicker than the older methods. When was the last time you went to the library, or opened an old-fashioned dictionary to look up a word's spelling? 

So how does that help us to value such services? Time is money. To get a very rough estimate of the value of your time consider how much you get paid per hour. Now consider how much time a day you spend searching online for information. Multiply the time spent searching by your hourly pay rate. Now double it. That's a rough estimate of how much online search engines benefit you financially. Or, if you're searching of information as part of your job, that's how much the technology benefits your employer by making you more productive.

But let's think about it from another perspective. Google saves us time. But is there not value from spending longer on the internet sometimes too? Think of time on social media catching up with friends and family. Erik Brynjolfsson of MIT conducted a survey in which Americans were asked how much they would have to be paid not to use the internet for a month. From these results he and colleagues were able to derive a rough estimate of how much value individuals derive from various online services. The value of social media - the likes of Facebook, Instagram and Snapchat - for Americans in 2017 was put at $322 a year. The value of search engines was put at a whopping $17,500.

Brynjolfsson presented his results at an Office for National Statistics conference about measuring the economy last week, hosted by the Bank of England. He argued that such benefits for consumers from the new digital economy are being missed in the national accounts of countries like the US and the UK - and that statisticians should consider using work like his to incorporate them. It wouldn't be a trivial adjustment. Brynjolfsson's figures suggest free sites added around $ 100bn a year to the US economy between 2007 and 2011, around 0.75 per cent of GDP.

Speaking at the same conference was Hal Varian, Google's in-house economic guru. Varian made a point about free cloud storage for digital photos, offered by firms like his employer. How much is this service worth? Well, he noted that Kodak used to say that people would rush into a burning home for three reasons - to retrieve family members, pets and, finally, photo albums. Again, the fact this claim came from a photography company might detract a bit from its credibility. But it rings true. The implication of the fact that many people would be willing to risk their very lives for their photos implies the digital photo revolution has an implicit financial value to people - and a significant one.

So should statisticians be making substantial adjustments to their national accounts as Brynjolfsson, Varian and many others suggest, in order to capture the undeniable value to consumers of new free online services? Aren't we underestimating an important area of economic growth by our failure to do this? There are reasons to be a bit cautious. First, as mentioned earlier, there is a question mark over whether we actually should regard all these services as free given some of these firms run a business model in which they sell our personal data to advertisers. Wikipedia might be a nonprofit, but Google and Facebook most certainly are not.

Second, there are many other goods beyond the digital realm that people value - and which one could attempt to measure through surveys - which do not enter the national accounts. Consider clean air, or friendship, or political freedom. How much would we pay not to choke on smog? How much would we pay to be free of the fear of being locked up by a repressive regime? Perhaps the problem is less that official measures such as GDP are not fit for purpose in the digital age but that, in the digital age, we place too much weight on GDP as an indicator of our changing quality of life.

Hammond’s spring statement was no new dawn for the UK economy. It was an optical illusion

The bonanza, in the end, failed to materialise. There had been a fair amount of talk before the Spring Statement that Philip Hammond would be in a position to unveil a new dawn for our beleaguered public finances. City analysts had been chattering like South London parakeets about a potential £10bn permanent improvement in the public finances.

And the excitement had jumped the species barrier from the Square Mile to Westminster. The ardent Brexiteer Jacob Rees-Mogg had squawked about the imperative of ratcheting up spending on the struggling health service. Other Tory MPs had similarly chirruped in favour of a loosening of the Government’s austerity corset. The Chancellor himself had allowed himself to refer to “light at the end of the tunnel” in conversation with Robert Peston at the weekend.

Well the light of the tunnel may not quite have been the headlights of an oncoming train, but, according to the Office for Budget Responsibility, it was certainly an optical illusion.

The OBR did revise down its estimate of public borrowing in 2017-18 by around £5bn on the back of fuller than expected tax revenues so far this fiscal year. Yet Robert Chote and his team concluded that this was not likely to be a permanent improvement, mainly because they now estimate that the economy is overheated relative to their estimates in November. In other words, when things naturally cool down, the tax revenues will start undershooting. 

Similarly, the overall GDP growth forecast for the UK is a smidgen better this year. But next year – when Big Ben (may) ring out for Brexit– it is unchanged at a paltry 1.3 per cent. This, we should bear in mind, at a time when the rest of the world is growing at its strongest rate in years. And the expected UK growth rate is actually shaved down in the final years of the OBR’s forecast in 2021 and 2022. What the Spring Statement giveth, it also taketh away.

The Chancellor might have rebranded himself as Tigger, rather than Eeyore, but the OBR was channelling the lugubrious blue donkey when it concluded: “There seems little reason to change our view of [the UK’s] medium-term growth potential.

City scribblers earn their bread speculating about the short-term ups and downs of the economy and the public finances. But what really matters for our long-term living standards is the UK’s productivity growth potential, the amount of output we can collectively squeeze out of each hour worked and each worker. And the big picture is that productivity has been as flat as a pancake ever since the Great Recession a decade ago. Or at least that was true until the second half of last year, when we, surprisingly, witnessed the best two consecutive quarters of productivity growth since the financial crisis.

Yet the OBR doesn’t think the UK’s productivity growth engine has started roaring again, since the improvement in 2017 was driven by fewer hours being worked, rather than higher output. A similar surge happened in 2011 before petering out. Another broken promisein other words. Let’s hope they are wrong, but there’s not much reason to believe it, particularly given ongoing public spending cuts and Brexit continue to dampen the animal spirits of households and companies.

And what of that pachyderm in the living room? What about Brexit? The consensus of serious economists is, of course, that Brexit will harm our growth potential by throwing up trade obstacles between us and ourbiggest commercial partners and, probably, reducing useful immigration to boot. And the OBR is not demurring from that. Its view, articulated way back in 2016, that Brexit would damage the UK’s public finances, wiping out any “dividend” from not making annual contributions to the EU Budget, still stands. And that’s based on the assumption the Brexit process is nice and smooth.

In the short-term, the OBR, conceded the economy had held up better than it expected immediately after the plebiscite. But this was partly due to the unexpected global growth spurt, partly because households seem to have dipped into their savings to support their spending. Moreover, as the OBR stressed today, it simply doesn’t regard the Office for National Statistics’ estimates of how the economy performed since the vote as reliable.

In short: as you were. The new Brexit dawn for the British economy looks remarkably like the old one.

Be wary of attempts to rewrite the recent economic history books

"The past isn't dead. It isn't even past," wrote William Faulkner. That certainly applies to the economy.

The Office for National Statistics (ONS) said last week that it had been examining its estimate for telecoms output-price inflation between 2010 and 2015. Research suggests statisticians may have been considerably overestimating this in light of large increases in bandwidth volumes available to business customers over that period.

This tentative and pretty technical bit of work has been seized upon in some quarters as implying a huge rewriting of recent UK economic history.

The BlackRock fund manager and former special advisor to George Osborne, Rupert Harrison, concluded that: "A lot of what we thought we knew about the economy is probably wrong."

Following up, Harrison's ex-boss himself tweeted that: "Official statistics underestimated growth for 2010-2015 and overestimated inflation, and therefore real income growth was higher during that period. Good to hear, although would have been nice to know at the time…."

The Times in an editorial joined the revisionist party stating: "The result is likely to have been an understatement of GDP and an overstatement of inflation."

A cynic might argue that it would be convenient for Osborne and Harrison to push the idea that when they were in charge of the British economy, GDP and income growth was much healthier than currently shown in the official statistics.

But leaving aside the motivations, are the conclusions that are being drawn valid? And the answer is that there's something a bit previous about them, to put it mildly.

First some perspective. Telecoms account for less than 2 per cent of GDP. That's not negligible, but it's not big enough to fundamentally change the picture of the economy, even if the sector really did grow substantially faster in real terms over the period in question.

And, in any case, aggregate output figures are always "balanced" by the ONS to aggregate expenditure calculations of GDP. In simple terms, this means that even if telecoms output is judged to have been stronger, output in other sectors of the economy would automatically have to be revised down to compensate. That's why Richard Heys of the ONS himself has stressed this telecoms prices revision, if it were implemented into the national accounts, would be very unlikely to affect the overall GDP figures.

But what about consumer welfare? Doesn't this work also suggest more bandwidth for ordinary phone and broadband customers too, giving them more bang for each buck of their expenditure? Hasn't consumer price inflation therefore been exaggerated too, and real income growth correspondingly underestimated? Again, we shouldn't jump to conclusions. First, telephone services and equipment account for 2.5 per cent of the UK's representative consumption basket. Again, that's not really big enough to have the kind of dramatic impact on the overall inflation rate that some of the revisionists are implying.

Further, the ONS's new work applies only to business-to-business telecoms services, not customer purchases of mobile phone data. And there is already an adjustment for rising quality for things like mobile phone tariffs in the Consumer Price Index. This adjustment might turn out to be too small, but it's misleading to imply there currently isn't any allowance at all for the fact that customers could be getting more for their money due to technological advances.

The issue of how statisticians account for the rise of the fast-growing digital economy is certainly a very important one. And there may well be GDP revisions ahead, perhaps even consumer inflation revisions, which have hitherto been extremely rare. The picture of the economy in the past is actually changing all the time due to regular methodological revisions in a whole host of areas by the ONS. The past really isn't dead.

Yet at the same time it's spurious to make bold assertions based on thin evidence and perhaps wishful thinking about what the impact is going to be. There's also an irony about the former Conservative Treasury team claiming some vindication for their economic strategy on the basis of this latest research, given the core rationale for their controversial austerity programme was not rising household incomes, or even GDP growth, but rapid reductions in public borrowing and the overall government deficit.

Recall that in the 2010 "emergency" Budget, Osborne outlined bold plans to achieve a surplus on the structural current budget in 2014-15. In fact it was still 2.6 per cent of GDP in that year. And there is still no projected overall surplus in sight, despite the many promises of ministers. The theory that the economy has been doing much better than implied by the current official figures sits rather uncomfortably with the fact that tax revenues have been so disappointing relative to expectations eight years ago.

The past may not be dead when it comes to official economic statistics, but that doesn't mean we can casually rewrite history in a way that suits our interests and reputations.