Not All Sun and Sand: A New Study on Lessons Learnt from Operating IFMIS Systems in the Pacific

Posted by Suhas Joshi, PFM Resident Advisor at the IMF’s Pacific Financial Technical Assistance Center (PFTAC)

Pacific countries with Integrated Financial Management Information System (IFMIS) systems have recently had a large dose of inputs, including several studies and a “lessons learnt” document. In 2008, following the PIFMA conference, several Pacific countries requested PFTAC’s assistance in assessing their FMIS systems. In addition to assessing their systems, countries sought suggestions regarding options for moving forward, either by improving existing systems or suggesting development of more modern systems. As a result of these requests, PFTAC initiated studies in six Pacific Island Countries (PICs) between November 2008 and December 2009. These studies covered the Cook Islands, Kiribati, Tonga, Tuvalu, Vanuatu and the Republic of Marshall Islands. These reports showed that similarities existed between various systems around the Pacific. The most important discovery was that there was a near identical FMIS technology stack (ie DB, OS, and hardware), and a common database architecture, in 5 of the 6 countries studied by PFTAC. Of course, there were several problems too--the use of inconsistent data networks, irregular reactive meetings between users, lack of a comprehensive plan for e-governance and little integration with Banks--the last not merely because in many Pacific countries like Nauru, Niue, and Tuvalu no real banking services exist.

As mentioned in my previous post, the lessons learnt document was presented to the 5th annual meeting of the Pacific Islands Financial Managers Association (PIFMA), recently held at Port Vila, Vanuatu. The document recognizes the long time taken to implement IFMIS and the lessons learnt from experience around the world. It, therefore, covers broad guidelines for the successful implementation of IFMIS projects in small island countries, on how to manage and sustain operational systems in small island states, as well as the lessons learnt from the study of the six Pacific Island Countries.

A fundamental choice in acquiring computer software is whether to "buy" or "build." The paper discusses the various options. The type of FMIS software found to be operating in the six countries covered by earlier PFTAC studies is Commercial Off-The-Shelf (COTS), or packaged software. Packaged software usually represents a more practical and sustainable solution for small countries. Using a packaged software approach helps keep system design issues to a minimum, thereby contributing to a more manageable implementation. The paper points out that this is especially important in resource constrained situations as are common among small Pacific island countries.

Another fundamental choice in acquiring computer software is whether to select commercial or "open source"  (OS) software. OS software is typically supported by non-profit entities including universities or industry user groups. Because the software is not marketed by a profit driven enterprise, licensing costs may be much less than for comparable commercial software. However, since OS software is not actively marketed, like COTS software, there are no product representatives to respond to tenders or provide technical support, although commercial system integrators may perform these functions in some cases. The tradeoff between COTS and open source software is the greater ongoing need for local technical expertise to manage, operate, and support open source software something not easily available in PICs.

In making a decision about which type of software to buy, a business case approach is important to evaluate and justify the choices. The paper maintains that in the context of small Pacific Island States choosing the type of software to buy should be a business decision. Cost should be a part of the information considered, but other factors including viability, risk, strategic relevance, appropriateness for regional or individual country circumstances, and ready availability of trained, talented technical support staff should also be considered. Ready availability of support resources is especially relevant in the Pacific. The paper then lists some suggestions for selecting IFMIS software and provides a quick reference guide to key features of the software seen in operation in the countries covered in the study.

The paper also highlights the importance of the use of on-line portals to train staff to ensure user involvement. In cases where the Information and Communication Technology (ICT) infrastructure needed for an IFMIS is in place, that same ICT infrastructure should also be used to train, inform, and motivate the users and stakeholders of the government-wide IFMIS. An online PFM information "portal" can be readily added to government-wide ICT infrastructure as an inexpensive incremental add-on capability. The paper emphasis the point that well-maintained IFMIS systems require far more resources than merely the initial capital costs of setting up such systems to function. It therefore emphasizes the need for proper handing over of projects from the supplier and installers to the on-going maintenance team and then invest in that maintenance to assist in this task.

Systems implemented primarily by off-shore "expert" consultants are not good candidates for successful sustainability. The report recommends that to help ensure that an IFMIS can be successfully operated and administered once it has been implemented, it is essential that key host nation staff be directly involved throughout IFMIS implementation and develop a sense of ownership towards the system if it is to be sustainable. This seems not to have been the case with many IFMIS implementations in the Pacific, where the transition from implementation to operations and maintenance has not always been smooth and seamless.

It is important to have a smooth transition from development to operations and maintenance (O&M). A technique to help smooth the transition from development, often under the technical leadership of offshore experts, to operations and maintenance (O&M), often under the technical leadership of local country staff, is to assemble the local team early in the requirements definition phase and be sure they understand the responsibilities they will be assuming in the future. Often, local staff do not realize that they will be assuming operational responsibility for the system in the future, or, if they do know, they may not have the background and experience to fully understand what will be required of them. The report addresses these issues as well as those relating to vendor management.

A separate 2008 PFTAC report[1] on Internal Audit in the Pacific, as well as the six country-specific PFTAC IFMIS reviews that form the basis of the lessons learnt document, all find that the piecemeal approach used in the Pacific for PFM information system development and management has not generally been very successful at delivering integrated PFM systems. The report recognizes that some good work has been done, some excellent progress has been achieved, and some valuable lessons have been learned. However, progress has, in most cases, been limited to implementation of stand alone FMISs having little or no integration with revenue, budget, planning, debt management, banking, and other PFM functions. In addition, most countries continue to struggle with network infrastructure issues associated with deployment of FMIS functionality to line ministries and departments. Overall, the piecemeal results achieved so far have not been as good as originally hoped for.

The report concludes that that there can be real and tangible benefits to be realized from regional coordination in this area, especially in the Pacific. The high degree of commonality among technology platforms potentially lends itself to the development of near universal FMIS related capabilities that could then be shared across multiple Pacific island countries at relatively low cost. Multi-country, common use capabilities, and documentation could be developed regionally in a coordinated effort using pooled resources from multiple countries, or it could be done by one country acting as a center of excellence and then shared with other interested countries as needed, or some combination of both these approaches. This could include a standard suite of ICT network devices capable of ready adaptation to specific country circumstances through a regionally coordinated research and development (R&D) effort with the resulting specifications, training, and even turnkey solutions being made available at low cost to interested countries.

To download the study, please click on the following link: Download IFMIS_PICspdf.

 

--------------------------------------------------------------------------------

[1] S. Joshi, Eroni Vatukola, Rakesh Gupta, An Evaluation of Internal Audit in Pacific Countries: The Way Forward. PFTAC, November 2008.

Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.

Comments

[Doug Hadden]

Very good work on analyzing IFMIS. These lessons are very important and should be considered by all developing nation governments. I'd like to add a few points.

Vendor Relationships

It's clear that vendor relations continues to be a key barrier to success in IFMIS implementations. As a vendor representative, it seems to me that the remedies suggested in the paper will only represent an incremental improvement. Vendors must understand the unique challenges in developing countries, including capacity and the need for constant PFM reform. Anecdotal evidence suggests that governments with a direct relationship with the software vendor tend to have more successful implementations than those who leverage the integration channel exclusively. And, developing nation governments do not have much leverage with most COTS vendors for product enhancements.

Legislation Reform

The impact of IFMIS on reform or reform on IFMIS is an important problem. There can be a loss of momentum when the IFMIS implementation has to wait for legal reform. Nevertheless, a core IFMIS can operate in most countries with little or no legal reform. The key problem is that the IFMIS must be able to adapt in line with legal reform. Simple is better, as indicated in the report. However, governments will progressively activate more sophisticated features. And, the Chart of Accounts will undergo changes over time to support modernization.

Maintaining IFMIS

As pointed out in the report, sustainable IFMIS is an important challenge. Countries should be self-sustaining. The use of higly priced foreign consultants should be the exception rather than the rule when it comes to post-implementation operational issues like problems or upgrades. Capacity building needs to be a fundamental deliverable by the COTS IFMIS vendor. Vendor support needs to go beyond the typical ICT good practices.

Clarification on FreeBalance features

I'd also like to provide some Annex 2 clarifications on FreeBalance software.

Cashiering: available in next release
Workflow: Yes, workflow management is provided
Maximum Number of Concurrent Users in implementation today: 5,000
Maximum Number of Concurrent Users: maximum has yet to be reached, confirm over 10,000
Works with MS Office: most recent version integrates with MS-Excel, MS-Word, Outlook
Operating System: Windows & Linux
Databases: MS-SQL,Oracle, MySQL
Report Writers: JasperSoft, Cognos, Crystal, Proprietary

Although JFMIP does not include budget preparation as "core", we are observing more and more Requests for Proposal requiring this functionality.

[Suhas Joshi]

Good recognition of the issues, Doug. The problem is that the smaller the country, the lesser is its leverage with IFMIs providers. In the context of small island states, this is compounded by lack of capacity at nearly all levels. Challenges can be best met by greater comprehension of the local issues on part of the vendors themselves, and their own recognition that change will be incremental. This would, of course, mean a movement away from “selling” their product to a more realistic implementation of systems viable in specific country context. But would that be feasible in a high pressure sales environment, normally observed in the IFMIS arena?
On the legislation issue- the path to success is to implement core modules first- rarely does any law prevent that. An additional bonus is that once the core areas are implemented properly confidence in IFMIS systems grows. This also prevents limited resources from being dissipated in implementing bells and whistles.
Overall, less pressure and hastening slowly would yield better long term results for all.