Correction to: Interest Groups & Advocacy https://doi.org/10.1057/s41309-019-00074-9

After the article was submitted for review, we continued improving the data. We performed a second attempt to merge groups in the lobby register with groups mentioned in ministerial meetings reports. We used Stata’s stnd_compname command to standardize names, followed by extensive eyes-on checking and internet searches where names were ambiguous. In this way, we found numerous duplicate groups in each dataset and across the datasets. In total, we identified 121 matches not reported in the article. Because we were critical of the lack of overlap between the two sources of data, we now correct the record, albeit without changing our original conclusions.

In Fig. 1 (using all available years), the updated data change: the number of groups that met with ministers from 22,295 to 21,046, the number of lobby clients and consultants from 1034 to 1041, and the overlap between the two from 452 to 684. Doing so alters the percent of lobbyists from the register that appear in the meetings data from 44 to 66% and changes the percent of groups in the meetings data that appear in the lobby register from 2 to 3%.

In Fig. 2 (covering just the period of overlap, i.e., 2015), the updated data change: the number of groups that met with ministers from 7303 to 6446, the number of lobby clients and consultants from 852 to 764, and the overlap between the two from 250 to 290. As a result of these improvements, the percent of lobbyists from the register that appear in the meetings data is now 38% (rather than 29) and the percent of groups in the meetings data that appear in the lobby register is now 4% (rather than 3). The now incorrect percentages are repeated several times in the article and abstract.

Counting in terms of meetings rather than groups, 87% (rather than 91) of reported ministerial meetings, or in 2015, 90% (rather than 91), were with groups and individuals whose names do not appear as clients in the lobby register. The newer data mean that the groups that appear in the meetings data that are also lobby clients account for 10% (rather than 8.6) of meetings, and the average lobby client appears in the meetings data 19 times (rather than 17).

The two-thirds of groups listed in the meetings data that attend just one meeting account for 19% (rather than 20) of the meetings. Meanwhile, the 10% most frequent visitors account for 64% (rather than 60) of all meetings with department officials, and the 1% most frequent visitors are present at 28% (rather than 27) of the meetings. Among ministry visitors that are also listed in the register, the top decile of most frequent visitors accounts for 58% (rather than 64) of the meetings, or 46% (rather than 45) in 2015 alone, and the top 1% are present at 15% (rather than 14) of all meetings with department officials, or 11% (rather than 8) in 2015 alone.

Finally, across the 5-year period, the meetings data include the names of 31 known consultant lobbyists by our count, while the same data contain 575 (rather than 431) registered lobby clients.

These improvements to the data notwithstanding, our conclusions about the inadequacy of both the lobby register and the ministerial meetings data remain unchanged.