By now, we have all heard about the sad hit-and-run killing of a female leopard in Yala recently. It is yet another instance of reckless driving by safari jeep drivers inside the national park, and another nail in the coffin of a wonderful park that is increasingly becoming too crowded to enjoy. In 2012, BBC’s correspondent in Sri Lanka Charles Haviland experienced this first hand, and pointed to “anarchic behavior by tourists and drivers” in his BBC article ‘Speeding and danger in Sri Lanka’s safari parks’. In June this year, the authorities announced a blocking of cell phone coverage inside the park, to prevent drivers from speeding around the park after being tipped off about whereabouts of leopards and other wildlife by other drivers.

Having being a regular visitor to Yala, I now avoid it, and prefer to go to the less crowded Wilpattu National Park instead, where drivers seem to be better behaved and the park is less congested. The scenes of the dead Yala female leopard were very saddening and I couldn’t help but wonder why not enough is being done about it. Explanations include “oh the jeep drivers are a mafia, none can control them; “they are all linked to powerful politicians so nothing can be changed”; “it’s too late now…”, and so on. To be sure, money is not a problem to introducing new measures. Yala is a very busy, but also a very rich park; it earned over Rs.274 million in park entrance fees alone in 2013.Yala Park  (3)

Here are some ideas on what can be done to tackle the problem, by influencing jeep driver incentives and thus, their behavior. Ideas 1 and 2 draw from regulatory economics principles, 3 from big data, and 4 from behavioral science.

1. Licensing regime: A stricter registration and licensing process for all safari jeeps and jeep drivers permitted to take tourists around the park. This permit is linked to the performance of the driver and the jeep. If the driver is found to be driving recklessly, endangering the passengers or the wildlife, the license could be revoked after a warning and probation period. If the vehicle is found to be air or noise polluting (heavy emissions or a noisy engine, above stipulated thresholds, for instance), the license could be revoked after a warning and an opportunity to rectify. A system of periodic review (linked to item 3 below) could be put in place, maybe every 6 months, where the driver and jeep owner have to get the license renewed.

2. Scarcity and rationing by introducing a limit on the number of permits in circulation: Let’s say for instance, that only 200 permits can be in circulation at any one time, once all permits have been given out for a year and a new entrant wants to get in, he/she must buy it off an existing permit-holder, who is allowed to sell it. Essentially, the principle is similar to that adopted in tradable pollution permits, in environmental economics. However, this process must be carefully managed to ensure that those with political patronage don’t get undue preference, and are granted multiple permits.

How permits are granted in year ‘t’ is also tricky – will all current drivers be granted permits? Will there be a ceiling on the total number of permits available/issued and drivers must bid for them? How equitable will that be and will it be politically feasible? Can the government consider reducing the number of permits in circulation by year t+5, t+10, etc., to try and get to a lower number of jeeps (either by buying them back, or by natural attrition – drivers leaving the industry for other jobs or old age, and his permit not being given out and being removed from circulation instead)?

3. Big data analytics by introducing GPS tracking for all licensed jeeps with licenses: Given that there is great telco coverage inside the park, a telco like Dialog can be contracted to provide jeep drivers with low-end smartphones with GPS capability, and the data can be analysed later, but also in real time. The real time tracking could help park officials determine which routes seem to be getting overcrowded with vehicles, and instruct subsequent entrants to the park to take alternative routes. The real time data can also alert trackers/rangers to ‘hot spots’ where vehicles are gathering, and dispatch officers from nearby posts to ensure the wildlife that is attracting the crowds to that spot is not being disturbed.

The data, logged with a unique number for each driver/jeep, can also be analysed later/periodically to assess the average speeds of the drivers and check for reckless driving patterns, and these could also feed in to the licensing process. To see how well this works, it can be introduced on a pilot set of 50 jeeps first. Moreover, if analysed alongside radio-collar tracking and camera trap data (are these there on any leopards, or only elephants?), could this GPS data help park authorities see how vehicle movement patterns are influencing wildlife behavior?

4. Influencing behavior by introducing an ‘Wildlife Friendly Driver’ program: Drivers will undergo a voluntary training module that introduces them to principles of being a better safari driver, respecting the animals and their natural habitat, etc., and subsequent to that, they will be given a certificate and an attractively designed sticker that they can display on their jeep.

Simultaneously, tourists must be made aware to look out for jeep drivers who have this sticker, are advised to make these ones their first choice and are (informally) “nudged” to reward them with higher ‘tips’. Once some of the drivers start doing it, tourists start making active choices, and other jeep drivers notice, then the “meter tuk tuk” effect would kick-in – other drivers would also want to ‘opt in’ to the scheme as they see extra value. A media campaign across the hotels around national parks, and around the park entrance premises, must give wide publicity to the Wildlife-Friendly Driver scheme. Tourists can be asked to fill a quick exit form to give feedback on their driver, and this information would contribute towards a bi-annual review of the driver’s certification.

Just some ideas.

While Yala is where these changes are needed the most, maybe we can start with Wilpattu where there’ll be less resistance, amidst a less Mafia’d scenario, and can demonstrate quicker wins?
A few caveats. i) I have only considered the safari jeep drivers operating from the local area, not the issue of private vehicles and ‘vandana’ busses that are now ubiquitous in the park. Maybe the solution is to ban them from entering the park, and only permit licensed vehicles? ii) I haven’t discussed how to influence behavior of drivers by altering behavior of the tourists/passengers.

Yala jeep driver behavior is surely a direct consequence of the generous tips that tourists (especially foreigners) give drivers if they have a successful leopard sighting. Drivers are naturally influenced to drive with the sole objective of “showing” a leopard. iii) I haven’t considered rationing the number of visitors to the park, as a way of reducing pressure on the parks resources. Like in Denali National Park in Alaska or even in tiger parks in India, there can be a rationing of the number of entrants each hour and each day.

Maybe some of the four ideas have already been thought of, or are being rolled out, but these are some running ideas I had. Would love to hear from those of you who travel to Yala more regularly and know the park more intimately than I do, and also those who have travelled on safari to Africa and have transferable lessons from there. Let me know what you think – what have I not considered? Which of these ideas are more feasible than others, given time, resource, and political constraints?

(Courtesy: The Curionomist Blog at

Yala Park  (4) Yala Park  (2)