Table 1.2 Economic Sectors (2017) – Other Measures (USD millions)
| Market Value | Profit | Revenue | Asset | Employees (total number) | |
| Manufacturing | 3,885,168 | 155,430 | 2,350,529 | 3,496,514 | 5,479,283 |
| % of Total | 17.99 | 15.47 | 18.34 | 8.30 | 19.41 |
| Financial | 3,400,038 | 228,481 | 2,232,306 | 27,909,063 | 3,198,268 |
| % of Total | 15.75 | 22.74 | 17.42 | 66.26 | 11.33 |
| Retail | 1,462,153 | 77,704 | 2,641,316 | 1,184,934 | 7,933,952 |
| % of Total | 6.77 | 7.73 | 20.61 | 2.81 | 28.10 |
| Service | 4,329,598 | 181,381 | 2,144,832 | 2,826,420 | 6,538,487 |
| % of Total | 20.05 | 18.05 | 16.74 | 6.71 | 23.16 |
| Digital | 6,650,784 | 276,929 | 2,047,313 | 3,821,682 | 4,254,320 |
| % of Total | 30.80 | 27.56 | 15.98 | 9.07 | 15.07 |
| Extractive | 1,865,371 | 84,876 | 1,398,194 | 2,879,857 | 827,529 |
| % of Total | 8.64 | 8.45 | 10.91 | 6.84 | 2.93 |
| Total | 21,593,112 | 1,004,802 | 12,814,490 | 42,118,470 | 28,231,839 |
It may be that such differences are indications of different kinds of economic activity, and it is tempting to immediately connect these findings to existing discussions about the digital economy. For example, to what extent is its very strong profit–asset relationship due to digital companies disintermediating government regulations and so avoiding asset costs that others must incur? Likewise, while the financial sector’s relation between profit and employee numbers may have an intuitive basis in the intangibility and large size of the commodities it deals with – shares, futures, derivatives and so on – it is tempting to connect the digital sector’s relatively moderate employment and high profit to the way many digital products benefit from extensive ‘free’ labour provided by the users of those products. Before being seduced by such numbers, however, it is important to face up to the more fundamental difficulty – which these and the OECD analyses both point to – of coming up with a clear definition of the digital economy.
Numbers do not speak. It is only through a conceptual framework that they appear to talk, even if we often forget that and instead take numbers as the gift-horse whose mouth it would be impolite to examine. From one perspective, the evidence suggests there is a digital sector of significant size, around 20 or perhaps even 30 per cent of the total economy, with a distinct pattern, while from the OECD perspective it is much smaller. These differences raise the question of what has been done to the numbers to get them to say these things? For example, while the FT and Fortune figures can be manipulated to produce the same sectoral division of the economy, the variation between them suggests that different definitions may be affecting the results, given the dramatic changes between 2015 and 2017. The definition of the digital economy as a statistical category resulted from the allocation of companies to particular sectors, the principles of which have not been clearly articulated. There is then no point in seeking more precise statistical results, or trusting anything but the most general results, until the definition of the digital economy is clearer.
Digital Economic Practices and the Problem
