Unfinished Business: Reforming Our Intelligence Agencies
Peter Jennings
Click here for PDF version

Terrorism is not the only urgent issue our intelligence agencies must address.

The last few years have been a particularly testing time for Australia’s intelligence agencies. Scarred by what was widely judged to be an ‘intelligence failure’ over weapons of mass destruction in Iraq, the agencies have been struggling with the even bigger challenge of reorganising to deal with terrorism. Al-Qaeda’s attacks in September 2001 redefined much of Australia’s strategic landscape. As a result our intelligence gathering and analysis capabilities have been pushed more prominently into the front line of national security. Spending on intelligence has in broad terms doubled since late 2001—now over $500 million is spent annually on foreign intelligence analysis and collection. For our premier analytical agency, the Office of National Assessments (ONA), spending will double again over the next three years.
Not surprisingly this rate of growth presents its own problems for the agencies’ capacity to recruit, train and retain staff. There are also some difficult decisions that need to be taken about the right way to structure the intelligence community as a whole and about how to set analytical priorities. In this article I focus on a number of these issues. Getting the answers right is far from being a trivial matter. If Australia is to detect and prevent the next terrorist atrocity, or to make the right judgement calls over issues of war and peace, then the starting point must be to have the best possible intelligence. Based on the record of the last few years, there is clearly room for improvement.

What follows reacts to the July 2004 Report of the Inquiry into Australian Intelligence Agencies. Commissioned by the Government in the wake of the Bali bombings and the Iraq war, the report was produced by Mr Philip Flood, a former Director-General of ONA and former Secretary of the Department of Foreign Affairs and Trade. When the report was released, the Prime Minister said the Government would implement all its recommendations (save for changing ONA’s name). It has therefore become the blueprint for reforming our intelligence system.

In many respects Mr Flood’s report is a model of the analysis one would like to see from the intelligence agencies. For the most part it is thorough and clear in its argument and it proposes many necessary reforms. It is also extremely cautious—another intelligence trait—and in my view does not go far enough with its recommendations for change in a number of critical areas. On one matter—the role of the Defence Intelligence Organisation—Mr Flood’s ecommendations could have a decidedly negative effect. Hopefully the Flood report will be seen as the start rather than the conclusion of a thoroughgoing effort to reform our intelligence agencies.

The intelligence community today

Australia has three agencies that collect overseas intelligence, the Defence Signals Directorate (DSD), the Defence Imagery and Geospatial Organisation (DIGO) and the Australian Secret Intelligence Service (ASIS)—one might say the ears, eyes and hands respectively of information gathering. Two assessment agencies, the Office of National Assessments (ONA) and the Defence Intelligence Organisation (DIO), use gathered intelligence as well as publicly available material to produce the finished assessments read by the Prime Minister and senior policy makers. There is also the Australian Security Intelligence Organisation (ASIO), which has a charter both to gather and to analyse information relating to Australia’s internal security from espionage, sabotage and politically motivated violence.

These agencies are often collectively referred to as the intelligence ‘community’ although what is most striking is the very high degree of autonomy each has to run their own affairs—autonomy far greater than any other government agency enjoys. The Director-General of ONA has an ambiguously defined responsibility to advise the Government on the adequacy and coordination of Australia’s foreign intelligence activities. But as Mr Flood notes, this function ‘has suffered’ over the years as the pressure of international events forced ONA to concentrate on its own reporting and less on community-wide issues.

The agencies tend to build strong, rather inward-looking work habits, structured around their highly specialised operations. Many positive consequences flow as a result, but one negative outcome is that the level of cooperation and understanding between the agencies tends to be thin, although this has improved in recent years. A huge effort has been put into building agency links in the area of counter-terrorism because an ‘all-source’ intelligence approach is necessary to defeat this threat. Generally though the agencies form a community in much the way that Metternich famously regarded Italy—as a geographic expression rather than a cohesive entity. Many of Mr Flood’s recommendations exhort the agencies to better understand each others’ roles and capacities.

A second essential feature of the intelligence agencies is that they are small. In the mid-1990s the Australian Defence Force ran a hearts and minds campaign against personnel cuts with the argument that a full-time force of 50,000 would be just half a good-sized Melbourne Cricket Ground crowd. Even after its recent considerable expansion, ONA struggles to fill a Canberra restaurant. It had 39 analysts on staff in mid-2004. In the lead-up to the Gulf War in early 2003, ONA could muster only three analysts to work on that issue. Here we can see the real cause of Australia’s intelligence problems on Iraq’s weapons of mass destruction. There is no need for conspiracy theories about manipulating politicians. ONA was not subverted, but simply swamped with information.

Until recently the relatively small size of our agencies has given them more scope to handle management issues informally. Inter-agency matters could be addressed at a monthly Heads of Intelligence Agencies Meeting. ONA in particular recruited largely from the best and brightest of the rest of the bureaucracy. On-the-job training, as Mr Flood kindly describes it, was ‘minimalist and ad hoc.’

The biggest challenge faced by the intelligence agencies today is that this ‘boutique’ approach will not work for organisations that have grown so quickly and so substantially. ONA’s annual budget was a tiny $6.29 million in 1999 and Mr Flood rightly proposes it should grow to $25 million by 2007. The ASIS budget of $44.6 million in 2000 doubled to $88.5 million this year. All the agencies, barring DIO, have been through similar growth spurts and, like any expanding business, they need new management structures to control their operations. In any business surely the most important structural questions are about the roles and responsibilities of the Chief Executive Officer. Yet, as we have seen, there is no CEO position in the Australian intelligence community.

Wanted: A Director-General of Intelligence

Perhaps because he was a former Director-General of ONA, Mr Flood treads delicately around the question of whether the existing, largely informal inter-agency top management structures are adequate. He points to the ambiguity that exists in the ONA Act about the Director-General’s responsibilities for coordinating and finding ways to improve the intelligence community’s work. Mr Flood judges that this task ‘. . . has not been fulfilled optimally in recent years’. ONA’s irector-General is also responsible for reporting to the Government his or her assessment of how well the agencies—including ONA itself—performs, clearly ‘a weakness in the system of performance reporting’, Mr Flood concludes.

The proposed solution is to formalise somewhat the current agency heads’ monthly meeting by creating a Foreign Intelligence Coordination Committee. The Director-General of ONA will get some extra resources to manage inter-agency ‘monitoring and coordinating’ and would report on these matters up to the National Security Committee of Cabinet. Mr Flood proposes that reporting on the adequacy of the intelligence agency’s performance should be shifted to the Department of Prime Minister and Cabinet. This latter step may be welcomed by ONA but, especially after the recent Iraq experience, it carries a dangerous potential to confuse the quality and accuracy of intelligence with the utility of that reporting to policy makers.

It is difficult to believe that these very limited measures will give the head of ONA the real authority necessary to make decisions on behalf of the whole intelligence community. They do not fix the problem inherent in asking one person to be a lead player at the same time as umpiring all the other competitors. That is an unenviable task and, as Flood’s report shows, Directors-General of ONA have tended to concentrate on their agency roles rather than their responsibilities for the whole intelligence community.

Mr Flood does not say whether he considered proposing a position of head of the intelligence community separate to the Director-General of ONA, but this does seem now to be a necessary change. For predictable bureaucratic reasons the idea may not be welcomed by existing agency heads, but the change offers many advantages.

A Director-General of Intelligence (DGI) could be given a specific charter to deal with a range of community-wide issues currently left in the too-hard basket. These include recruitment strategies across the agencies rather than the current habit of mutual poaching of staff; developing a community-wide approach to the growth of ‘open-source’ public information and working out how best to set collection and reporting priorities across agencies. The DGI would occupy a genuinely independent position from which to report to Government on the agencies’ performance. The position could take a large share of the increasing demand for agency heads to make public appearances—giving evidence, for example, before the proliferating committees of backbench parliamentarians. Importantly, the DGI’s position would allow ONA’s Director-General to concentrate on intelligence analysis, surely no bad thing.

To function effectively, the DGI will need a small supporting staff, perhaps in the order of twelve to twenty people. At that size it is clear the DGI will not be carrying out intelligence assessment functions, but the position will have the capacity to help shape key intelligence judgements, particularly in areas where agencies may have different interpretations. In its formative stages, though, the DGI’s main role will be to manage issues that are consequential across the intelligence community.

In time the absence of a Director-General of Intelligence will come to be seen as bizarre as the idea of the Army, Navy and Air Force operating without a Chief of Defence Force—as they did twenty years ago. The sooner we create this unifying leadership position the better.

An Intelligence College

An early priority for the first DGI should be to determine which elements of recruitment, training and career development ought to be managed commonly across the agencies. The Flood report identifies this as a crucial issue, going to the heart of the intelligence community’s ability to attract and retain the best talent. There will certainly be clear agreement on the areas of training that shouldn’t be centralised—that is the highly specialised tradecraft of the individual collection agencies. There is, however, an urgent need to develop common induction training, build solid analytical skills and to create more in-depth opportunities for analysts and collectors to hone their skills.

Readers will be disappointed to learn from Mr Flood’s report that, in ONA ‘…there is little to no conscious management of staff’s training or development needs.’ Across the community there is no common induction training, few advanced training opportunities for senior staff, no agreement on the training needed to assure competence and no framework for career progression. Until recently ONA saw training as an unnecessary expense. They recruited skilled staff from elsewhere and hoped that learning on-the-job would add essential specialist skills.
This was never a satisfactory situation, but it is now unsustainable given the agencies’ demand for increased numbers. Mr Flood’s recommendations in this area are welcome, but they fall far short of what is really needed. He calls for a community-wide orientation and training course, some common training for the analytical agencies and ‘a programme of strategic seminars’ for senior leaders. But disappointingly, Mr Flood argues that the gap can be filled by ‘a two person secretariat’ to manage this training effort. Such a minimal commitment falls way below what is needed and, it must be said, way below the expectations of the young and skilled graduates the agencies most need. For many young people, a key factor determining where they choose to work is the opportunity for further professional training and education. In a buoyant economy this is a seller’s market and unless the intelligence agencies can compete by offering career paths with the chance for further professional development, they will fail to attract sufficient numbers of the most talented people.

Establishing an Intelligence College outside of Defence’s military intelligence training would help to solve this immediate recruitment problem. It would provide a lasting focus for deepening Australia’s capacity to undertake complex intelligence analysis. Salted throughout Mr Flood’s report are hints and indications where the agencies have performed poorly because the system lacks depth of analysis and experience. ONA is said to be ‘…literally only “one deep” in expertise’ on any given topic, the agencies suffer from a ‘real skill deficit’, they do not understand each other’s strengths and limitations and are forced to borrow staff from other departments to sustain expertise. There is insufficient knowledge of foreign languages and not enough opportunities for overseas learning. In short, Mr Flood describes a system chronically starved of skills and living off the intellectual resources of a few senior officials.

During the election campaign, the Government committed $20 million to establish what it called a ‘Centre for Counter Terrorism Intelligence Cooperation and Joint Training Centre for Australia, South East Asia and the Pacific.’ This looks promising, but it conflates a number of tasks that need separate consideration. Regional cooperation is important, for example, but not at the expense of the unique training needs of Australian intelligence personnel. Counter-terrorism cooperation is also important, but it is far from the only strategic issue on which our agencies must report. The election proposal seems to be a clever adjunct to an Intelligence College rather than a substitute for it.

None of the agencies’ current training difficulties can be fixed overnight, but an Intelligence College would be an enormously important start. The College would need resources sufficient to run community-wide training for around a hundred or more recruits every year. It would need to offer a range of programmes designed to deepen skills as people become more experienced and to provide a means for collectors and analysts to understand the vital roles each play. A faculty of around a dozen or more intelligence specialists with a lively training programme might cost in the order of $5 million annually—even by intelligence standards a small investment. Given the annual $500 million spent on foreign intelligence, it is nothing short of astounding that the community has not yet found it possible to allocate one per cent of that sum to common training.

Matters of content

In recent years the overwhelmingly large part of the assessment agencies workload has been devoted to producing ‘current intelligence’ reports—fast-breaking information relating to the crisis of the moment. Obviously this is a critical part of intelligence, but the Flood report is concerned that short-term reporting is crowding out the capacity to produce long term assessments. These make judgements about strategic trends over a decade or more. They are a vital foundation for strategic thinking, helping policy-makers sort out the significant from the trivial.

Mr Flood also charts the decline of the National Assessment—special studies produced across the intelligence community with ONA managing the process. In its early years, ONA produced these by the dozen—39 in 1979 for example, considerably more than have been written over the last ten years. National Assessments are important for a number of reasons. They integrate the views of all the agencies, they provide an agreed analytic foundation for decision-makers and they give context to short term reporting. National Assessments are also hard to produce, time-consuming and require serious thinking to prepare.

The Flood report is right to call on ONA to produce both more long-term assessments and National Assessments. The expansion of the agency’s numbers should make that possible, at least in time, if more experienced staff are freed from the current intelligence treadmill to do this long-term work. At a time of fundamental and rapid strategic change it makes sense to revisit these judgements often and never to be too confident about the accuracy of one’s conclusions. But governments have no choice other than to rely on this type of analysis as a guide for policy decision-making. Across the broader public service, long term strategic planning has declined as increasing workloads squeeze departments, but in the intelligence agencies at least this should be treated as vital work.

Another area of increasing importance to the intelligence community is the explosion of unclassified information—known as ‘open source’ material in the trade. Intelligence professionals can at times under rate open-source information but the massive expansion of the Internet alone makes this a potentially vital information source. We have seen, for example, the growing use of the Internet by radical Islamic terrorist groups. Such web sites require constant monitoring. Mr Flood has recommended the transfer of an Open-Source Unit from the Department of Foreign Affairs and Trade to ONA. This is aimed at more closely integrating open source and intelligence material and seems a sensible step. But at an annual cost of $2.5 million the Open Source Unit can hardly be scratching the surface of the volume of information that is both available and of potential intelligence value. Over time it is inevitable that additional investment will be needed in this field.

Dealing with DIO

Finally we come to one of the more puzzling recommendations in Mr Flood’s report—on the always-contentious issue of how to manage the relationship between ONA and its military counterpart, DIO. A certain amount of creative tension has always marked relations between ONA and DIO. DIO has clear responsibilities to produce intelligence assessments needed by the Defence Force, especially when it deploys on military operations. But there is a substantial overlap between ONA and DIO on the issues they study—perhaps up to 75% of reporting from the two agencies covers the same subject matter, albeit from different perspectives. Differences of interpretation can lead to heated subterranean disputes between analysts.

One such difference was made public by the December 2003 Parliamentary Committee report, Intelligence on Iraq’s Weapons of Mass Destruction, which tracked a growing ‘difference of emphasis’ in ONA and DIO’s reporting in the lead-up to war. The report found DIO’s views ‘…more sceptical and circumspect than those of ONA in the same period’ about Iraq’s holding of, and capacity to use, chemical and biological weapons.

How should one deal with these differences of opinion when such judgements are clearly vital to policy makers? It is important for differences to be aired—for ‘contestability of assessments’ as it is called, but policy makers have some right to expect that intelligence judgements will be clear enough to help them make decisions, not just to mire the process with debate and uncertainty. As Mr Flood weighs these issues, he makes some useful recommendations that will improve matters. He calls on ONA and DIO to make more strenuous efforts to sort out key differences of judgement, and he says that politicians must be alerted to the points where interpretations diverge.

At the end, though, Mr Flood’s view is clear: DIO ‘. . . should cease publishing intelligence not directly serving requirements for strategic-level military-related analysis. The resultant product should be more strongly defence oriented and distributed primarily to defence customers.’ So much for the contestability of assessments! Australia would not be well served by taking this step. There is no clear dividing line between intelligence serving national needs and ‘strategic-level military-related analysis.’ If this injunction was applied to ONA, then large areas of its published output would be blank. I see no alternative other than to encourage ONA and DIO to continue reporting in overlapping areas. We need stronger processes pushing analysts to resolve differences—and the independent position of a Director-General of Intelligence would help. Ultimately, though, the expression of different strategic judgements is a vital part of the intelligence business. It is as well to remember that, on Iraq at least, DIO’s assessments were shown to be the more accurate. That effort shouldn’t be rewarded now by narrowing the scope of DIO’s reporting.

Conclusion: a continuing reform agenda

The growing threat of terrorism has been the key driving force behind the massive expansion of Australia’s intelligence agencies since 2001. However this growth is unlikely to be limited to the spending currently planned or the mostly very necessary recommendations detailed in the Flood report. Much of this current growth has focussed on immediate needs—covering gaps in short-term reporting and responding to current international crises. But there is a deeper need to address some fundamental questions about the intelligence community’s top management structures, the way it recruits and trains people and the balance between long and short-term reporting. Seen this way, the Flood report should not end the business of reform in our intelligence system. Indeed it is only a starting point for a continuing task likely to bring even more fundamental change to this key component of Australia’s national security.

Peter Jennings is Acting Director of the Australian Strategic Policy Institute in Canberra. These are his personal views.

Policy is the quarterly review of The Centre for Independent Studies. For more information on subscribing to Policy, click HERE

If you are interested in the Centre's activities and publications, why not subscribe to e-PreCIS, our regular email update on the latest news and events.

(e-PreCIS requires html capable email facilities, such as Microsoft Outlook Express or Netscape Messenger)