Skip to content

Seeing is not believing: how people see corruption in the UK

Image by Lonpicman
Image by Lonpicman

How can we tell how serious the problem of public service corruption is in any given country? If you live in the UK, you might – if we believe Transparency International’s new research – think that such corruption is a significant problem. Only 2% of the people questioned said that corruption was ‘not a problem at all’ in the UK public service, whilst 29% of people said it was a ‘very serious problem’.

Conversely, public perceptions are much more positive in Yemen and South Sudan than they are in the UK. In Yemen, 7% of the public said corruption in the public sector was ‘not a problem at all’, whilst only 11% said it was a ‘very serious problem’. In South Sudan, 29% said corruption was ‘not a problem at all’, and only 11% said it was ‘a very serious problem’. So, if you went only by such perceptions, you might conclude that South Sudan and Yemen were notably less corrupt than the UK.

Of course, perceptions do not necessarily reflect reality, and it could be that these particular perceptions are seriously misleading. Indeed, according to several expert assessments, Yemen is one of the world’s most corrupt countries, and – whilst South Sudan is too new to have much formalised data about it – it appears that South Sudan is also probably one of the most corrupt countries on earth.

So why are perceptions so negative in the UK? When you look closely at the data, there is one standout figure that might offer some clue. Of those in the UK who have had some interaction with the judiciary, or where a member of their household has, 21% report a bribe being paid to a member of a judiciary. This is twice the rate claimed in South Sudan, although still half of what is claimed in Yemen. On this aspect of the public sector alone, it would appear that there is a genuine problem with corruption. Given the importance of the judiciary for maintaining integrity in general, and given the power they wield, this level of corruption could, and indeed rightly should, result in very negative perceptions.

Except, there is a problem with this analysis – it runs directly counter to other available evidence. On corruption in the judiciary in the UK, the Council of Europe’s Group of States Against Corruption (GRECO) said as recently as March 2013 that:

‘The judiciary is ranked as the most trusted institution by the public in the United Kingdom, with an untarnished reputation of independence, impartiality and integrity of its members. Nothing that emerged from the current evaluation indicated that there was any element of corruption in relation to judges nor was there any evidence of their decisions being influenced in an inappropriate manner.’

So, again, we are left wondering why people in the UK are so negative, and why it appears from the figures in the latest Global Corruption Barometer that there is a significant bribery problem in the UK judiciary.  As with all surveys, we need to explore the underlying data in more detail.  This particular survey was conducted with a sample of 1000 UK residents online. Online surveys are great for many things, but we know that they can be a poor way to estimate the extent of a phenomenon in a society. In Yemen the interviews were face-to-face. Perhaps most puzzlingly, in South Sudan the interviews were conducted by telephone. South Sudan essentially has no land-line capacity, and its mobile penetration was estimated in 2011 to be around 13% (even if likely to be expanding quickly). Given this, it is highly likely the South Sudanese survey respondents were considerably better off than the average citizen of South Sudan.

Of course, crucially, the results do not mean that 21% of people in the UK have experienced judicial corruption in the last year. Indeed, as the results only apply to the subset of people who have had personal or familial interactions with the judiciary in the last year, the ‘21%’ is actually only 15 people (21% of 7% of 1000). This is closer to the roughly ‘0’ we might have expected, but likely still too high given what we know about judicial integrity from other sources.

Yet, if the situation is not so bad in the UK, and yet perceptions are so hostile relative to countries that – by expert assessment – seem to be performing much worse, what does this mean for the quality of the data collected? At the moment, the best advice we could give would be to be highly cautious about using the data, and be very aware how others are using it.

Paul Heywood and Jonathan Rose

Published inCorruption


  1. Richard Rose and C. Peiffer are trying to control for that assymetry of the sensitivity of the responses by differentiating between “perception of corruption” and actual person experience. Controling for the latter allows to improve the significance and efficience of the corruption indexes.

    For a full discussion, see R. Rose and C. Peiffer (2012), Paying Bribes to Get Public
    Services: a Global Guide to Concepts and Survey Measures. Glasgow: U. Of Strathclyde in Studies Studies in Public Policy (SPP) series No. 494

    • Jonathan Rose Jonathan Rose

      George, that’s a really interesting line that they’re taking (although unfortunately I can’t see the whole thing, as it’s paywalled), but one of the things that we’re seeing with the data discussed here is a complete disconnect between perceptions, reported experience, and (probably) reality.

      Obviously, we can’t measure the reality of bribe paying, so we don’t ever really know the ‘true’ rate – but we can often build some pretty solid expectations based upon other data. So if we take the UK (which is the country I know best), I simply don’t buy that 21% of people who have had contact with the judiciary have paid a bribe in the last year. Yet that’s what people here seem to report as their actual experiences. It’s doubly odd, because the general figures for people in the UK having ever paid a bribe are only around 2% – a figure that, ironically, comes from a 2010 report by TI themselves (see–part-one—national-opinion-survey , p.4). Or, outside of quantitative surveys, TI also say “There is little evidence that corruption in the UK legal profession is prevalent and recorded incidences are very low” (–part-two—assessment-of-key-sectors , p.24).

      TI, of course, claims that the amount of bribe paying has simply shot up – but to go from 1.7% reporting ever paying a bribe to around 5 % saying they had done so in the last year… I’m not buying that. It simply goes against everything we know outside of that one survey that was used for the GCB. Methodologically, as best as can be told from the information available from TI, the 2010 report was far better – a random sample of more people, inherently taking in a far wider range of people.

      So my issue is – yes – with the disconnect between experience and perceptions, but also (perhaps especially) between reported experience in this case and with what we can infer reality to be. This will be a far more difficult task to address.

      For what it’s worth, here’s my guess as to what caused this: 15 people were caught breaking the law for which the punishment is a fine levied through the magistrates court. I’m guessing speeding, or some other relatively low-level motoring offence. They were incensed that they – a law abiding citizen who just happened to break the law – could ever be required to receive any form of punishment when there are ‘real criminals’ (who may or may not have broken the law) around. They see this as so unjust that they regard them paying a legally levied fine as a bribe. Obviously, I don’t know if this is the case, but I could easily imagine it being so.

      I think TI need to be far more careful with pinning down what they mean when they ask the bribe paying question again (they should continue to ask about contact also, and I don’t have too many problems with how they did that). I would advise them to use something far more like:
      “Bribery occurs when a person illegally pays another person in order to alter the actions or decisions of the person being paid. Have you paid a bribe in the last twelve months to any member of the following professions in order to influence their professional work…”
      OK – at the moment it’s maybe no so elegant, but I think it needs to be something far more like this

Leave a Reply

Your email address will not be published.