THERE two ways to see yesterday’s cabinet reshuffle by Rishi Sunak, which saw Suella Braverman get rid of as home secretary and David Cameron return to the front line of British politics. It’s either the death knell of a tired government or a master stroke of lateral thinking.
Whatever the interpretation, it used to be a major cabinet reshuffle – which we define as more than four cabinet ministers moving in one day, or any of the chancellor, home secretary or foreign secretary – being rather rare occasions. In the 30 years before the Brexit referendum vote in 2016 they happened about once every two years. But in the chaos that followed, a renegotiation has taken place every six months on average. Constant churning at the top of government is bad for policy continuity. Does renegotiation have any effect on voters?
To find out, The Economist reorganized yesterday, using a database of government ministries maintained by the Government Institute, a think tank. We ran data going back to Margaret Thatcher’s government (which began in 1979) with polling figures, to see how party positions changed in the eight weeks before and after each major reshuffle.
Overall, we find that renegotiation has little effect on voters’ intentions. Some prime ministers seem to have been better at implementing cabinet changes, or at least timing them, than others (see chart). However, overall, the share of voters who intended to vote for the ruling party declined by an average of 0.3 percentage points in the eight weeks following major reshuffle, compared with a decline of 0.1 in before the cabinet reshuffle was announced. This should come as little surprise: cabinet changes are often not performed from a power position.
That is certainly true of Mr. Sunak’s shakedown. According to our poll tracker, the Conservative Party’s massive 21 percentage point deficit to Labor has been widening recently. The reorganization gained a lot of attention but also sparked a new round of Tory infighting. It is highly unlikely that the picture will change either. ■
