Establishing new markets where things can be priced and traded creates incentives for people and companies to work out how much something is worth to them and to trade and innovate accordingly. It is hoped that allowing us a greater ability to control access to our personal data, in exchange for services and money, could encourage similar benefits.[1]

Part of this may involve the growth of Personal Information Management Services (PIMS). PIMS are an overarching term for intermediary services which help individuals manage their data, and companies’ access to it.[2] It has, for example, been suggested that companies could pay fees to PIMS providers to access the data that users have chosen to make available, fees which consumers would share in.[3] Data that PIMS might cover includes browsing data, transactions data and health data, among others.

In principle, allowing greater transparency and control in the use of personal data should facilitate new forms of economic activity. There are, though, a few reasons to suspect that an image of individuals playing a very active role managing their own data will not be how this pans out. The limitations of this suggest some of the ways this area could evolve in future.

1. People can fail to do comparatively simple things that are in their economic interest

There is evidence that consumers, even when presented with choices that would benefit them economically, may not be willing to engage. Just because people have opportunities to trade, it does not mean they will do so. The UK’s energy regulator Ofgem has, for example, estimated that the average savings from switching from the incumbent electricity and gas suppliers’ to the best available tariff is roughly £100 per year.[4] Yet the majority of users do not switch, and the price gap has remained roughly constant since 2008. Energy tariffs can have complications, but energy is ultimately a simple product; the electricity and gas sold by a company is physically indistinguishable from another.

2. Data is not a straightforward commodity

In contrast to electricity, data is a very heterogeneous commodity. One person’s data is not the same as another’s and it is not immediately clear what it should be worth. In addition, those who best understand the commercial value of data are not the individuals who generate it, but the companies that already have a lot of it, and devote extensive resources to analysing it. The limited degree to which people currently study the complicated terms and conditions associated with their data shows some of the challenges in engaging them. Albeit that greater standardisation encouraged by a more developed market for personal data will hopefully streamline these.

3. Much of the commercial value of personal data does not exist at individual level, but is created from analysing it alongside other people’s at scale

An individual’s personal data in isolation provides information about what they have done previously, but it is not necessarily that informative about new things they might like or that might happen to them (data on other people helps work that out). As a result, personal data is worth more when combined with data from many other people. From this it also follows that it is the analytical processing of the data that creates a lot of the data’s value, rather than the individual’s data itself, and the value of the individual data is probably limited.

In addition, in the context of advertising, a lot of adverts’ value typically comes from being able to show adverts to a large number of people, and there is some empirical evidence that suggests the value of being able to target adverts is comparatively low.[5]

4. There is already a secondary market for personal data

A secondary, or perhaps more accurately, second-hand, market for personal data already exists - although most people whose data is traded on it are unaware of this.[6][7] Data brokers collect publicly available information about us and then resell that information to companies. It has been estimated that the main purposes for this are to inform marketing (e.g. to target advertising and inform customer segmentation) and to allow companies to assess risk (e.g. credit, insurance and fraud risk).[8] There can be a wide range of potential sources of this data from public records, data from browser cookies, information on social media activity available via Applications Programming Interfaces (APIs). There is also evidence that certain mobile apps extract extensive personal information about us.[9] The existence of a lot of accessible information about us already, and the ease with which it can be copied, puts constraints on the economic value of the data that PIMS allow us to control.

This is not to argue against the principle of individuals having greater transparency and control over their data, or to disagree that there are benefits from creating new incentives in this area. It also does not do justice to the rich variety of different kinds of data, and to the distinct purposes it can be used for. It does, though, suggest that people's motivations and choices are likely to be subject to some constraints.

This implies a few ways that the increasingly sophisticated management of our personal data may go:

• PIMS will make decisions for us (or at least suggest them)
  
  It seems likely that there will be an element of PIMS making decisions for us. From 1, 2, and 3 above, it follows that it is unlikely that most people will want to be actively managing their data, and even if they did it might be better if someone else (or an algorithm) did it on their behalf - subject to constraints set by the user. This implies that the PIMS may make decisions on behalf of customers, or at least do a lot of the legwork in terms of offering them the best deals to choose from.
   
• There will be consolidation amongst PIMS
  
  From 3 and 4 (and also the previous point), it follows that PIMs will benefit from economies of scale/market power in terms of being able to negotiate better deals on behalf of their customers. As noted by the Competition Markets Authority (CMA), a challenge is that PIMS are a two-sided market, where in order for both firms to engage with PIMS, a critical mass of people has to be using them, and vice versa – which should encourage consolidation. [10] Scale would allow PIMS greater ability to undertake more sophisticated analysis on behalf of consumers. Consolidation of PIMS is also more likely to help control data leakage into a secondary market, which would help to increase the data’s value and the sustainability of their business model. It is perhaps arguable that some of the existing internet platforms that people already share their extensive personal data with have an advantage in providing PIMS services in certain contexts.
   
• There will need to be clarity on whose behalf PIMS are operating, and where data is going
  
  Personal data can be used by companies to offer us better deals, but also worse ones too. Data might suggest someone is prepared to pay more for something, or that they are a worse risk. Part of establishing a market in this area will require PIMS to be explicit about their role in these transactions, and informing the user how their data is being used - could it, for example, affect their future credit rating or insurance premia?
   
• PIMS may provide one way to address some of the switching issues mentioned earlier
  
  There are a whole series of utility markets: electricity, water, banking etc. where consumers often would be better off if they switched provider, and yet they don’t. Many reasons have been given for this[11], but one of the underlying drivers is that a lot of people are just not interested in the process of buying these products. They buy them because they have to, not because they enjoy doing it. All of which should mean a role for third parties that are able to use people’s data, work out what the best deal for them is, and save them time and money.

Some readers of the above may consider it a call for specific regulation of the emerging PIMS sector. It is not. There should, overall, be greater attention to data protection issues, given the more central role that data is playing in our lives, but we should also be open to the possibility of new business models, particularly given the lack of transparency in current ones. Indeed, one of the key reasons for allowing experimentation is to uncover new things that can’t be learnt from existing arrangements. The economy of the future should not be completely constrained by the economics of the present.

Thanks go to Katja Bego for helpful comments on this post.

[1] World Economic Forum (2011),’Personal Data: The emergence of a new asset class’.

[2] Ctrl-Shift (2014), Personal Information Management Systems: An analysis of an emerging economy, Nesta.

[3] Ctrl-Shift (2015), ‘The data driven economy: Toward sustainable growth’.

[4] Office of Fair Trading, Competition and Markets Authority and Office of Gas and Electricity Markets. State of the Market Assessment 2014, p51.

[5] Farahat, A. and Bailey, M. (2012), How Effective is Targeted Advertising? WWW 2012 – Session: Advertising on the Web

[6] Federal Trade Commission (2014), ‘Data Brokers A Call for Transparency and Accountability’.

[7] The D-CENT project (2015), ‘Research on Identity Ecosystem’.

[8] Federal Trade Commission Op. cit. IV Type of products

[9] Liccardi, I., Pato, J. and Weitzner, D. (2013) ‘Improving Mobile App Selection through Transparency and Better Permissions Analysis’, Journal of Privacy and Confidentiality: Vol. 5: Issue 2, Article 1.

[10] CMA (2015), ‘The commercial use of consumer data’, p83.

[11] The UK Regulators (2014) report on Consumer Engagement and Switching lists 12 key barriers that prevent people switching.