Celebratory send-off lunch — Mark is towards far end of table.
Dear colleagues,
Professor Mark Horridge retired at the end of February 2025 and is now an Honorary Professor in the Centre of Policy Studies (CoPS) at Victoria University, Melbourne. Everyone at CoPS would like to acknowledge the enormous contribution Mark has made to CGE modelling and GEMPACK software over the last 40 years.
For the last 18 years Mark has been Professor and Director, GEMPACK software at Monash and Victoria Universities. Many in the CGE community will know Mark's work through the software he designed. Some of his programs are part of the GEMPACK suite, others are associated with GTAP and many are ancillary programs for data manipulation. Mark developed the first Windows, or GUI, programs for GEMPACK on the PC in the 90's, extending Ken Pearson's vision of developing software to enable those without specialist computing skills to run CGE models.
Mark introduced best practice to the coding of models. He introduced a systematic naming convention for variables and coefficients, now familiar to many. Mark's naming scheme and clean coding style have allowed models based on theory developed substantially by his long-term colleague Peter Dixon and others to become more accessible to a global community of users.
He is the creator of the path-breaking TERM approach to multi-regional sub-national modelling. This enables the user to create a much more detailed regional model than is possible by relying only on official national accounts data. Versions of TERM have been implemented for Austria, Brazil, Canada, China, Finland, Germany, Indonesia, Italy, Japan, Korea, New Zealand, Poland, South Africa, Sri Lanka, Sweden, USA and Vietnam. TERM has been used in more than 90 academic publications.
Mark created the original MMR model (later to become MMRF and now known as VURM), an eight-region bottom-up model of Australia. VURM has been the backbone of most of the CoPS Australian economic modelling work over the last two decades.
Mark has written many database calculation programs used to convert input-output or supply-use tables to a CGE format. He has been instrumental in extending the core theory of CoPS-based CGE models. Applications include multiple household extensions, top-down extensions and mapping US counties to congressional districts. Many of the materials prepared by Mark are available in the CoPS archive.
Mark's MINIMAL, ORANI-G, TERM and Database course materials have been used to conduct dozens of short courses over the past decades in Australia and around the world.
Mark has been associated with GTAP since its inception and his contribution to the GTAP community has been monumental. On the GTAP website, 78 GTAP-Related Resources appear under his name, covering an array of software including RunGTAP and GTAPAgg, and database and theoretical innovations.
Mark has shown extraordinary generosity to visiting academics and PhD candidates over several decades. As a keen traveller Mark has worked with many collaborators, too numerous to mention, around the world. Perhaps most notable is Brazil where he has regularly visited since the late 90's, working with Professor Joaquim Bento de Souza Ferreira Filho and Professor Edson Domingues amongst others.
Although retired, Mark plans to continue working much as usual.
We wish Mark the very best for his retirement.
Celebratory send-off lunch — Mark is towards far end of table.
At a lunch to celebrate his retirement, Mark Horridge gave a short speech:
After 42 years working at 3 universities in Melbourne, I discovered that my superannuation fund will pay me a tax-free lifetime annuity which slightly exceeds my post-tax salary from CoPS. So I chose to let my super fund support me, rather than CoPS. A side-benefit is that I no longer need to complete the University's annoying online training modules.
Following 6 years in the UK, in December 1981 I returned to Canberra on holiday — and decided to stay. To amuse myself on the plane, I had bought a Sharp pocket computer, with a 1-line screen and full keyboard, about the size of a mobile phone. On this I learnt BASIC, assembly language and machine code.
The main employer in Canberra was the Publice Service, which recruited via an annual entry exam — that I had just missed! I followed a neighbour's suggestion that, while waiting for the next exam, I complete a 1-year diploma in economics. Economics was very fashionable at that time, while computers were starting to appear everywhere. Both economics and computer skills were in high demand.
For the diploma entrance exam, 1st year Economics had be mastered, then the 2nd and 3rd years of an Economics degree were compressed into 1 year. During the year I also took a course in Fortran.
After the next Public Service exam, I was allocated to the IAC (Industries Assistance Commission, previously Tariff Board, later Productivity Commission). They seemed very suspicious of someone with two arts degrees. They asked if I would like to work in Melbourne with the great Professor Powell — I never heard of him. I said, no thanks. Then I was told that the alternative was to bend paperclips for 2 years in a very boring section. I agreed to go to Melbourne to work at the Impact Project, attached to Melbourne University. This proved to be a great opportunity and I have worked with Impact and its successors ever since. I was encouraged to enroll in an Economics PhD while working, which I completed late 1987.
My career coincided with a 30-year improvement in computing power. When I started at Impact, punched cards were still used to store programs and data for the mainframe computer. We decided to replace the cumbersome card punch machine with a rival of the first IBM PC: the NEC APC with a 5MHz 8086 CPU and up to 256 kilobytes RAM. We realized that the PC could also be used for word processing and for creating and running our own programs. We graduated to 286 IBM clone PCs which we used until 1991, when we moved to Monash. Then we jumped to 486/7 PCs, which ran 30 times faster! It was now possible to solve CGE models on the desktop, and we gradually stopped using the mainframe computers on which we had depended.
The same period saw the growth and maturation of CGE modelling. From a few pioneers fighting computational constraints, the field expanded all over the world. CGE models became a standard tool, especially for modelling trade issues, and later for simulating environmental issues.
I enjoyed 3 great features of working for Impact/CoPS: