Earning sentences

Last month saw the emergence of two studies of particular interest to those of us interested in health.

Executive pay

One concludes that there’s little evidence to support highly paid CEOs as agents of turnaround and change in the NHS. Where CEOs are medically qualified clinical care may be slightly better; where the hospital is in trouble, a financial sector background may help. But other than that the authors find little to support the big pay (and big pay differentials) seen amongst NHS CEOs.

This suggests that the very people who insist that clinicians submit business cases and show value for money are themselves not value for money.

This is a striking conclusion. If CEO pay differentials are not based on observable merit, it raises concern not only about waste, but also unfairness. The BBC recently lost a discrimination case, having paid a man six times what it paid a woman doing the equivalent work. It had tried to justify this by arguing her work was in fact qualitatively different. This argument crumbled under quantitative scrutiny by an employment tribunal. Any claimed difference in work proved intangible.

Further quantitative scrutiny of CEOs’ work should be done to determine what drives pay, if not performance. A tenable explanation may be that CEO pay is being driven by patronage. If this is also true of CEO recruitment, it may also help explain what’s been observed as “The Snowy White Peaks of the NHS”. In other words, CEOs may enjoy unearned privilege arising from the assumption they’re people of value.

An alternate explanation is that CEO pay is driven not by clinical performance but by other measures, such as political utility. CEOs may be useful to the Department of Health as lightning conductors when things go wrong. When children like Caitlyn Parry die avoidably at Alder Hey, it may be politically useful for CEO Louise Shepherd to claim all concerns about her case have been “proven absolutely untrue” while quietly compensating her mother because a Royal College of Surgeons report shows Shepherd was not telling the truth. Likewise in the whistleblowing case of Dr Chris Day, the Lewisham CEO may be useful to take flak, and divert it from more senior bureaucrats in Health Education England.

Even if this is the case, and CEOs are politically useful to those higher up the food chain, what determines the often eye watering levels of pay and pension? After all when things get sticky the CEOs often defer to expensive outside legal counsel to do their job for them. We can investigate such matters with data and in detail. But there may be a lack of appetite to do so.

Prosecutor bias

The second study from the USA shows the way the NHS might understand such issues much better. Indeed, it highlights the lack of research ingenuity and investment when looking at differential outcomes across the NHS – from BAME patients, to the attainment and disciplinary experiences of BAME staff.

The study was covered in The Economist. The researcher was interested in biases within the US criminal justice system. He found that when sentencing guidelines changed, prosecutors adjusted their charges – inflating the amount of crack cocaine said to be involved in the case. But they did this less for white offenders. Systematically, he explores for other explanations before showing that this problem is present in about a third of prosecutors and this misconduct is personal in that it follows them when they move to a formerly non-biased district. He shows too that their behaviour correlates with measures of societal prejudice. In states where this is low, prosecutors don’t misbehave like this. Where it’s high, they do. He finishes by calculating the consequences. The benefit to white offenders in costs and time served is huge. Racism has long been lucrative.

Assuming 3.6% and 1.2% of all drugcases from 1999-2015 were subject to similar discretion by race implies total costs of 1.3 billion dollars for black and Hispanic offenders versus 148 million dollars for white offenders. In terms of incarceration, the disparity implies 21,000 years sentenced due to this discretion for black and Hispanic offenders versus 2,500 years sentenced for white offenders.

– Racial Disparities in Federal Sentencing: Evidence from Drug Mandatory Minimums

This second study adds to the first in that it shows how much deeper one can delve into these issues. In time, we should apply such techniques to problems like differential attainment amongst BAME and white medics, and to the Snowy White Peaks. Only in that way will be begin to understand the extent of unearned privilege within our nation’s biggest employer.