The “Pacific Solution”

A cost blow-out whichever way the tide turns

 

1.  Executive Summary

This paper forecasts the costs of the Pacific Solution by building a model of the average cost per day per detainee.  It concludes that the daily costs of the Pacific Solution are higher than the costs of detention in Australia.  The costs range from $200 to $400 per day per detainee, compared to $120 in Australia and $67 if the refugees are released into the community.  This contradicts the Government’s statement on this[i]. 

The model forecasts that, if asylum seeker arrivals continue at about 4,000 per annum,  the numbers detained on the islands will rise from the current level  of 1,676[ii] people to a permanent population of over 4,000 refugees within five years. This will require significant additional infrastructure and likely extension of the program to other countries.  A second issue of concern is the length of time people will spend on the islands – the model predicts that within five years, there will be 1,300 people on the islands who have been held there for over two years.  The implications of such lengths of detention are now being felt in Australia.  The tragedies of recent months could be exported to Pacific islands.

 

2.  Background

In a previous paper I have analysed the cost of Australia’s internal detention policy[iii].  That paper concluded that the average cost of housing an asylum seeker in detention in Australia was $117 per person per day.  This has since been confirmed by DIMA who have stated that the cost per day is $120.[iv]

Since then the Government has moved to a new approach to asylum seekers called the “Pacific Solution”.  Under this scheme asylum seekers are flown to Nauru or to Papua New Guinea (Manus Island, a remote island near the equator).

I wanted to analyse the costs of the Pacific Solution to the tax payer, as well as to predict how “big” the island detention centres would become.   To do this, I constructed a “multi-decrement model” showing the build-up of numbers of detainees under the Pacific Solution.  The model forecasts both the number of asylum seekers being detained, the length of time of each detention and the average cost per day of detention at any point of time.  A multi decrement model is a common actuarial forecasting system used to model populations of lives over different durations.

As expected, the model is sensitive to three assumptions:

The amount of money spent on the Pacific Solution, either directly through constructing the detention camps, or indirectly through Government payouts.

The number of people attempting to seek asylum in Australia, and therefore placed in a detention camp on an island.  These are called “ new entrants” in the model.

The rate at which claims are processed and people are moved off the islands to either their country of destination or returned home.  This is called the “release rate”

Each of these are discussed below in Sections 3,4 and 5 of this report.  Section 6 shows base results and Section 7 shows a variety of scenarios on the base forecast.

 

3.  The amount invested in the Pacific Solution

Reports on the cost of the Pacific Solution vary wildly.  In September 2001 the Treasurer, Peter Costello, said the cost would be $103 million[v].  In January 2002, a leaked Cabinet paper said that the government had already spent $285 million on the Pacific Solution and was budgeting a further $200 million for the next  five years (ie $1,285 million in total).  Mr Ruddock refused to deny these figures, but went on to say that the cost of detaining these people in Australia could have been over $300 million[vi].

The Australian Financial Review then ran an article quoting Government sources as saying that the Department of Immigration (DIMA) and the Department of Defence were seeking additional funding of $1.8 billion to fund “the Coalition's policies on security and asylum seekers”[vii].  Only part of this is for asylum seekers but it gives a sense of the magnitude of the numbers.

These numbers may seem incredible to those aware that we are dealing with only 1,676 asylum seekers to date.  A detailed breakdown of the $285 million spent to date is not available, but costs in the public domain have included:

 

Initial cost components of the Pacific Solution[viii]

$m

Nauru facility cost

72

Nauru additional aid

30

Manus Island facility cost

24

PNG additional aid package

20

MV Tampa military costs

20

Christmas Island and Coco Island costs

8

Cost components known  to date

174

 

This paper uses the leaked Cabinet number of $285 million as the spend to date, given that a large number of known costs (flying the asylum seekers to the islands, military efforts subsequent to the Tampa) are not included in the table above. There are also significant investments in the islands’ infrastructure as the following quotes show. “Nauru's infrastructure problems should soon be a thing of the past”[ix] and “The development of the detention centre for 446 on Manus Island had brought many benefits to local business and an upgrade to water, power and sewerage works for the island's naval base”.[x]  It is easy to see how the $285 million could be reached. 

But what of the spend going forward?  The asylum seekers will be subject to the same expensive prison style regime that costs $120 per day in Australia.  In the absence of other information, we have used the leaked government figure of $200 million per annum as the cost if the inflows stay at 4,000 per annum.  If the inflows reduce, we have assumed a pro rata reduction in the annual costs, down to a fixed annual spend of $4 million while some detainees remain .

 

4.  The number of new entrants to the camps

The numbers housed on the islands at 31 December 2001and 5 February 2002, and the current capacity of the camps, is as follows:

 

Current numbers in detention

31 December 2001

Latest Estimate 7 Feb 2002

Current Capacity

Nauru

1,118

1,149

1,200

Manus Island - PNG

216

360

446

Christmas Island

211

34

200 est

Cocos Island

131

131

200 est

Total

1,676

1,676

2,046

 

After up to nearly 6 months in the camps, none of the detainees had been released (24 August 2001 to 7 February 2002, the date of this paper).

How many asylum seekers can we expect will be brought the islands?   The following table shows the number of refugees attempting to seek asylum in Australia via a boat.  These numbers have been constructed from monthly table of boat arrivals on the DIMA website[xi].

 

Number of asylum seekers (by boat)

 

June 1998 to June 1999

1,000

June 1999 to June 2000

4,174

June 2000 to June 2001

4,163

July 2001 to Feb 2002

3,208

 

As is known, the number of refugees seeking asylum has increased over the past two years to about 4,000 per annum.  However the number is not “leaping upwards”, month by month, as the Government claims.  Instead, the number of people arriving each month fluctuates wildly. The table below shows numbers of asylum seekers by month for the past 36 months. 

Numbers are jumping around.  They fall as well as rise.  The peak in November 1999 was higher than any subsequent peak.  Numbers fell to zero in July 2001.

 

For the purposes of the model, I have used 4,000 people as the base number of annual new entrants to the camps.  I have then run three scenarios:

The number of asylum seekers falls to zero (I have called this Scenario Howard)

The number of asylum seekers halves (I have called this Scenario Peace Time)

The numbers of asylum seekers rises by 10% per annum compound (I have called this Scenario War Time).

 

5.  Rate of release from detention

This is the real driver of the populations in the camps.    The rate of processing (both the initial assessment and the resolution of appeals) is difficult to get hard facts on.  DIMA have stated that 80% of all refugees are processed within the first 18 weeks[xii].   This appears to be contradicted by:

The time taken to process the current people on the Pacific Islands – as of February 7 2002, all detainees were still on the islands, being up to 6 months from their date of detention (MV Tampa came on August 24).  The majority of detainees have now been detained for over three months.

The current make-up of the population of detainees.  As at December 2000 58% of the detainees in camps had been there more than 3 months[xiii].  Given the flow rate of refugees around that period, this is inconsistent with the processing rates DIMA is quoting.

After studying the various DIMA and other materials available, I have constructed a base release rate from the centres as shown below.  Note that this release rate includes both initial processing and time for a decision following appeal:

Under my model, 70% of detainees are released within 6 months.  This is considerably better than DIMA is managing at the moment with the Pacific Solution, but presumably reflects DIMA’s aspirations better.  A key feature of the release statistics available is that the release rate slows significantly after the first six months.  This is due to the cases left becoming progressively more difficult to process and to the appeals process being much slower than the original assessment process.  Once detainees have been inside more than two years, their rate of release slows so much that they are very likely to end up detained for up to five years. 

 

Having derived opening numbers (1,676), new entrants (4,000 per annum) and release rates (as shown above), we can now forecast the numbers detained under the Pacific Solution as follows:

 

After five years, the numbers detained has increased to over 4,000.  As shown in Section 2, the camps currently only hold 2,046. This kind of scenario has presumably been factored into the Government’s forecast expenditure.  But of possibly even more importance is the number of people who have been detained for over two years.  By 2007, the model predicts 1,300 people in detention for over two years.  This means that the psychological and mental problems currently evident in Australia will become a problem for the Pacific Solution.  Other papers have dealt with these issues.

 

6.   Forecasts of daily costs per detainee under base scenario 

The Government has not invested $285 million just to house 1,676 people for a few months. It has invested this money as a long term “solution” to the asylum seeker arrivals.  But how does this investment compare to housing the detainees in Australia?

By dividing the cumulative amount spent by the Government on the Pacific Solution by the cumulative number of detainee days, we get an average “cost per detainee per day”.  On day one, the cost is high because a large investment has been made ($285 million) for a small number of detainee days (1,676 people for 3 to 6 months).  But over time the daily cost falls, as the capital expenditure is spread over more detainees and more days.

The base case daily cost is shown below:

 

 

What does this tell us?  It tells us that, even after five years and with 4,000 asylum seekers arriving per year, the Pacific Solution costs over $200 per person per day.  This high cost is because of the high up front cost (the $285 million that did not need to be spent) and the high ongoing costs (some $200 million per annum).  The local detention, while sending profits back to the US private prison company Wackenhut[xiv], would at least have provided some economic flow on and a daily cost of $120.  In either case, detention is much more expensive than release into the Australian community – a daily cost of $67.

 


7.  Forecast Scenarios

The figures above are based on a central estimate, but it is important to test them with some “what if” scenarios.   I have tested the model under three such scenarios:

The number of asylum seekers falls to zero (I have called this Scenario Howard).  Under this option, the annual costs of $200 million fall to $4 million.  The camps are maintained at current levels, additional aid stops and the remaining detainees are fed.

The number of asylum seekers halves (I have called this Scenario Peace Time).  Under this option, I have assumed that the annual costs are also halved to $100 million.

The numbers of asylum seekers rises by 10% per annum compound (I have called this Scenario War Time).  Under this option, the annual cost also rises by 10%.

The graph below shows the cumulative costs.  Interestingly, the worst financial outcome is Scenario Howard.  This is because the initial investment of $285 million has been spent on only 1,676 people for a some months.  This is a financial disaster compared to having continued detention in Australia.

The Howard Scenario has led to a Pacific Solution cost of over $400 per detainee per day.  All other scenarios lead to a cost of $200 plus per day.

Naomi Edwards

BSc (Hons) FIA FNZSA FIAA

 

Notes and sources:

Below..

Below…

 

 

 

 

 

 



[i] SMH Date: 25/01/2002

The Federal Government says that if it had not been for its so-called Pacific solution, more boat people would have come and the cost of detaining them would have been more than $300 million.

The Immigration Minister, Philip Ruddock, refused yesterday to deny figures published in the Herald, based, according to sources, on cabinet documents that suggested the Government had spent $285 million to date on its policy of sending asylum seekers to Pacific countries.

The source also claimed the documents said the Government was planning to budget $200 million a year over five years for its "solution". The Government estimated in October, during the election campaign, that it would spend just $103 million extra on the solution this financial year plus $45 million on capital costs.

Mr Ruddock said yesterday: "There are ongoing expenses associated with the Pacific solution, and it'd be naive to believe we're going to stop feeding people and we're going to stop appropriate provision. The costs have to be looked at in the context of the extent to which the Pacific solution is part of the overall deterrents that are in place."

At the time of the Tampa stand-off, about 3500 people had travelled in four months and this could have risen to 10,000 to 12,000 arrivals over 12 months. "If you want to compare costs, ask yourself what the costs of detention onshore in facilities here ... [were if you] accommodate larger numbers of people ... it was going to be over $300 million just in immediate detention costs alone if we had those sorts of numbers.''

[ii] DIMA Fact Sheet 76 Offshore Processing Arrangements

[iii] Financial Analysis of Detention Centre Costs, Naomi Edwards, 25 January 2002

[iv] DIMA Fact Sheet 73 Unlawful Arrivals

[v] SMH 24/01/2002 “During the election campaign, the Treasurer, Peter Costello, said that only an extra $103 million would be required for the solution, with another $45 million to build new detention centers”

[vi] See Note I above

[vii] Australian Financial Review “The Federal Government faces major new pressures on its Budget bottom line with the defence and immigration departments pushing for more than $1.8 billion in extra funding to pay for the Coalition's policies on security and asylum seekers.  The big new spending bids come as the Government yesterday canvassed changes to the operation of the strife-torn Woomera refugee detention centre and the Prime Minister, Mr John Howard, vowed to defend Australia's refugee policy during his visit to the United States.  But the Coalition's tough stance on refugees and Australia's response to the US terror attacks are believed to have triggered significant cost pressures in the defence and immigration portfolios. Government sources said the departments were seeking urgent Cabinet approval for a total of $590 million in extra funding for the rest of this financial year alone - an amount that would eliminate the estimated 2001-02 Budget surplus of $500 million.  The departments are also understood to have indicated that they will then push for a total of at least
$1.2 billion in extra funding in 2002-03. “

 

[viii] These numbers are sourced from DIMA Fact Sheet 76 and from Oxfam’s excellent paper on the topic at http://www.caa.org.au/campaigns/refugees/pacificsolution/partone.html

[ix] 17/01/2002 6:16:40 PM | ABC Radio Australia News,” Australia to assist Nauru with infrastructure rebuild:  Nauru's infrastructure problems should soon be a thing of the past, according to the island's former President, Kinza Clodumar.  Power was lost for several days after a fire last week, and all telecommunications links between Nauru and the outside world were cut earlier this week. Mr Clodumar, says under the deal reached with Canberra over taking asylum seekers, Australia has promised to assist the island cope with it's ongoing problems: "Australia has offered in its initial offer of last October to provide some infrastructure for the power generation, which they are doing, and some of our men from the phosphate industry are down in Melbourne looking at the hardware here before it is shipped. The impact will be much better infrastructure all round and I suppose Australia is looking at it on a longer term basis for the processing of refugees. There is rumours you know that it could be on a long-term basis."”

[x] 01 FEBRUARY 2002 MELBOURNE (Pacnews) “Professor Waiko said the development of the detention centre for 446 on Manus Island had brought "many benefits" to local business and an upgrade to water, power and sewerage works for the island's naval base. But Oxfam's Mr Hewett said financial inducements like the $30 million pledged to Nauru had skewed Australian aid priorities in the Pacific. The distortion and politicising of the aid program damaged Australia's reputation and added to regional instability.  Placing an extra 1000 asylum seekers on Nauru put extra strain on an already difficult situation.

[xi] DIMA Fact Sheet 74 Unauthorised Arrivals by Air and Sea

[xii] DIMA website “Rebuttals to misconceptions about detention” states “Cases are processed in the shortest possible time, with 80% of cases finalised in under 18 weeks. Many people choose to pursue negative decisions through appeal processes, thus prolonging their stay in immigration detention”.

 

[xiii] The Detention of Boat People by Adrienne Millbank,  Social Policy Group DIMA 27 February 2001 : “Measures were adopted in the mid-1990s to speed up the processing of refugee claims for boat people. However time is inevitably taken to establish identities, and to appeal unfavourable decisions through the Courts. At the end of December 2000, of 2023 people in detention 31 per cent had been held for less than one month, 20 per cent between one and three months, 13 per cent between three and six months, 18 per cent six to 12 months and 18 per cent for a year or more.”

[xiv] In the year to December 2001, Wackenhut had $98million in revenues from DIMA.  See my previous paper.