Contracts and convicts: How perverse incentives created the death fleet

It might sound strange, but the tragic voyage of the second fleet, which killed 40% of the convicts on board, holds a lesson for contemporary governments about the consequences of perverse incentives. Here The Mandarin's David Donaldson reports on the work of ANZSOG scholars Gary Sturgess, George Argyrous and Sara Rahman.

The outsourcing of public services is often seen as a relatively recent phenomenon, its genesis lying in the shift towards free market thinking in the 1980s and 90s.

But while it’s not widely recognised, Australia’s founding myth is tied up with outsourcing. As with many public services in the late-18th and 19th centuries, convicts brought to the Australian colonies were transported by private companies under contract to the British government.

Dilemmas faced by governments writing contracts for complex human services today, such as trade-offs between price and quality, and how to capture motives and incentives effectively, were present over 200 years ago.

Nowhere are the hazards of perverse incentives more obvious than in the notorious tale of the second fleet in 1790, which led to the deaths of 40% of the convicts on board, during the journey and in the months after landing. This was bad even by the low standards of the day, when the average mortality rate hovered around 5-10%.

Conditions on the voyage — run by the slave trading company Camden, Calvert & King — were horrific, with convicts being poorly treated and given inadequate provisions. Despite an outcry in London after details became public, the Navy Board did not impose any financial penalty on the contractor. So what went wrong?

[Image sourced from Mitchell Library, Library of New South Wales]

A well-funded journey to the end of the earth

The story starts with the first fleet, which made it to New South Wales with a mortality rate of 5.4% — a success when you consider it involved carrying several shiploads of people to an almost completely unknown land on the other side of the world.

The first fleet contract was primarily a “cost-plus” arrangement, also known as a cost reimbursement contract, where the contractor is paid for all of its allowed expenses up to a set limit.

Outsourcing expert Gary Sturgess says that “given the uncertainty about the length of the voyage and the cost of provisions to be purchased at foreign ports, it is not difficult to understand why such a funding mechanism was adopted.”

The government — and convicts — were lucky in the choice of firm, too. The contractor, William Richards, hoped that by doing a good enough job he would win future contracts in this new area of government activity, and spent more on obtaining quality provisions and unfunded requests from the government than was required by the contract. Although his hopes ultimately turned out to be misplaced, “the prospect of repeat business encouraged Richards to invest heavily in building capability,” explains Sturgess.

Richards seems to have had personal motivations for doing a decent job, too — he was a humanitarian, a devout Christian and likely an abolitionist, telling prime minister William Pitt that prison management contracts needed to be based on “proper care and humanity to those unfortunate wretches”.

In addition, governor Arthur Philip, who was on the journey and more concerned with the success of the journey than cost, acted as a monitor of contract performance.

The cost-plus model made the first fleet expensive, however, notching up the equivalent of around AUD$110 million in today’s currency.

The lowest price is not always right

Outrage among Treasury penny-pinchers at the bill from the first fleet coincided with the arrival of a new cost-cutting home secretary, the 29-year old Lord Grenville. So it was decided the second fleet would be set up on a fixed-price per head basis, with the contract going to the lowest bidder. The amount decided upon was much less than the first fleet, promising a cheaper solution to the problem of overflowing gaols at home.

But payment was not dependent on how many convicts reached the other side. Most companies were unwilling to take on that risk, because at the time low standards meant death rates were high and often influenced by chance and the weather.

Instead, the contractor was paid for the number of convicts who boarded the ship in England. They were also allowed to sell any provisions left over after the journey, which they did.

This not only failed to create an incentive to treat prisoners reasonably, but meant the company made more profit if fewer survived.

In the hands of a slave-trading company, built on the misery of other humans, the results were disastrous. Of the approximately 1000 prisoners who sailed from Portsmouth, 267 died at sea and at least 150 soon after reaching Sydney.

William Hill, a second captain in the New South Wales Corps who sailed with the fleet, thought the conditions found in the slave trade compared well to the second fleet. The masters of two of the ships, the Neptune and the Scarborough, were cruel, he wrote:

“The more they can withhold from the unhappy wretches, the more provisions they have to dispose of on a foreign market, and the earlier in the voyage they die the longer they can draw the deceased’s allowance for themselves; for I fear few of them are honest enough to make a just return of the dates of their deaths to their employers.”

Of the 499 who boarded the Neptune — the worst ship — only 72 arrived in decent health. 158 died; 269 were incapacitated at the time of landing. The Reverend Richard Johnson described the scene soon after the fleet arrived thus:

“The misery I saw amongst them is indescribable … their heads, bodies, clothes, blankets, were all full of lice. They were wretched, naked, filthy, dirty, lousy, and many of them utterly unable to stand, to creep, or even to stir hand or foot.”

The contract imposed heavy fines for escapes, so troublesome convicts were placed in leg restraints designed for the slave trade, which, Hill said, prevented movement at the risk of a broken leg.

There had been no provision for stopping along the way written into the contract, so there was no incentive for the company to do so, despite the fact that the fresh provisions would have improved the health of those on board. They stopped just once, at Cape Town. The result was that the second fleet was much faster than the first fleet (which made three stops), arriving in 160 days, rather than 250. Again, this saved the company money, but increased suffering for the passengers.

The shock of the incident prompted talk of criminal charges, but the company was never indicted — and had already been contracted for the third fleet before the news became public. And although a strict inquiry was recommended by the attorney general, government concerns about exposing further details of the problems with transportation meant it never happened.

Lessons

A paper on the history of convict contracting in the December edition of the Australian Journal of Public Administration highlights the importance of system design, effective monitoring and management when public services are delivered under contract.

Although the world has changed dramatically since then, many of the contracting problems faced by government in the colonial era persist, argue Gary Sturgess, George Argyrous and Sara Rahman of the Australia and New Zealand School of Government.

The path taken to reduce death rates over several decades is instructive. Following the early voyages, changes to contract incentives and the introduction of an oversight role led to improvements in conditions on ships, bringing down the mortality rate from the early years average of 5-10% to around 2-3% during 1801-1815. Strengthened oversight and other gradual improvements helped bring it down further, eventually to around 1%.

The problems with the second fleet arose as the British government was moving from the traditional, relational approach to contracting — where government reuses companies and negotiates prices with them — to a more transactional one, based on the lowest cost regardless of factors such as experience and capacity.

This change reduced opportunities for corruption and political decision-making, but it also decreased incentives for tenderers to do a good job. The authors argue that this new model was not appropriate for such a difficult job:

“Complex social services cannot be commissioned or delivered in that way. There was a need for contractors to understand the human dimensions of convict management, and the transactional, price-based system meant that new providers had little relevant knowledge or capability and little interest in developing such capabilities, requiring government to play a much greater role to reduce mortality rates.”

Regularly changing contractors meant firms did not build up knowledge about how to deliver services well. This was remedied when the government eventually appointed surgeon and former prisons inspector Jeremiah Fitzpatrick to develop, formalise, and institutionalise best practice.

The capacity for learning and continuous improvement forms an important part of such a regime, argue Sturgess et al. If the government does not design the system to ensure companies develop this capacity, it must take on the responsibility itself:

“… Every public service contracting system, particularly those involving services to the vulnerable, needs a Jeremiah Fitzpatrick: someone who is authorised to be troublesome in taking up the cause of end-users, someone with a mandate to call providers (public or private) back to the fundamental outcomes, someone with a deep understanding of the challenges involved in front-line delivery and what must change to ensure that service objectives are met. This is the role of the public service commissioner.”

Later contracts were designed not only to give the surgeon the powers to make changes to increase safety and quality of life, but a financial bonus based on the number of convicts landed alive. This led to significant changes in the regime, “including the introduction of routine health inspections prior to the convicts being sent on board from the hulks or prisons, changes to the layout of the convict quarters, improved sanitation, ventilation and water filtration, and a requirement that the official table of rations be posted in the prison (making it easier for the convicts to protest about their conditions).”

So while transportation may have ended, fitting contract design to the context is still vital, particularly when it comes to difficult services, Sturgess, Argyrous and Rahman remind us:

“… This is an important lesson: contractual form should follow service function. Contract models suitable for buying pencils and light bulbs will not work particularly well in procuring the delivery of complex human services.”

---

Posted by @MsSophieRae