progress.
Key to my proposal is the phrase “self-enforcing.” It is an arrangement that requires little or no cooperation or trust between Israelis and Palestinians. The idea I have in mind provides each side with incentives to promote peace and resist terrorism purely in their own interest and utterly without concern for whether it helps the other side. In that sense it follows game theory’s dismal view of human nature.
My idea is that the Israeli and Palestinian governments will distribute a portion of their tax revenue generated from tourism (and only from tourism) to each other. Before going into the details—where the devil resides—let’s first see why tourist revenue and not any other. Why not, for example, promote peace by setting up joint Israeli- Palestinian ventures, or allowing freer movement between regions, or some other scheme? As we will see, shared tourist revenue provides a nearly unique opportunity.
The Palestinian Authority (PA) leadership routinely identifies tourism as one of the major pillars of the future Palestine’s economy. This is an eminently reasonable expectation given the vast number of historic and religiously important sites within the current and expected future territory of Palestine. The PA’s gross domestic product in 2007 was $4.8 billion. During peaceful periods, tourism represents more than 10 percent of income, and it could be much, much higher. By comparison, Israel’s GDP in 2007 was over $160 billion. Its tourist revenue was $2.9 billion in 2005 and $2.8 billion in 2006, and it is expected to be around $4.2 billion in 2008. So tourist revenue is a nice but relatively modest source of Israel’s income.
Tourism has a feature that can be exploited to improve the prospects of peace. You see, tourism and the tax revenue generated from it are highly sensitive to violence. For example, take a look at figure 7.1. The horizontal axis shows the range of violent Palestinian and Israeli deaths resulting from their conflict for the years from 1988 until 2002.2 The scale reflects a range of quarterly deaths from 0 to about 300. The vertical axis shows the number of tourists (in thousands) who visited Israel each quarter between the same period, 1988 to 2002.3 Unfortunately, I have not located comparable tourist data for the PA, but I have found enough to see that the pattern is the same. When violence goes up, tourism goes down, and when violence drops, tourism returns.
The line in the figure shows the estimated rate of tourist response three months after the reported level of violence, while the dots show the observed amount of tourism in Israel associated with actual violence that occurred three months earlier. Tourism is delayed by three months after observed violence to give prospective visitors enough time to change their plans.
FIG. 7.1. How Much Does Israeli Tourism Respond to Violence?
Clearly, more violence means a lot fewer tourists. In fact, on average, every violent death translates into 1,300 fewer tourists and 2,550 fewer hotel bed-nights sold to tourists. There were 53 violent deaths during the typical quarter covered here. That translates into nearly 70,000 fewer tourists in a three-month period suffering average violence compared to the number expected in quarters with no dispute-related violent deaths. Israel averages about 450,000 tourists per quarter, so 70,000 is a serious number. With Israel enjoying about $3 billion in tourist revenue in an average recent year, and an average year having about 280,000 fewer tourists than might conservatively be expected in a peaceful year, that translates into about $500 million tourist dollars lost each year, not counting whatever is also lost by hotels, restaurants, taxis, car rental companies, guides, and so forth on the Palestinian side of the border.
And what is the Palestinian experience like? As I mentioned, it is harder to get equivalent data for the PA, but still there is plenty of evidence that the picture is grim. For instance, there were about 90 hotels in the PA before the intifada that started in late 2000. By the end of 2001 the number had dropped precipitously to around 75. Naturally, hotel openings and closings result from how much business they do. Figure 7.2 shows the number of hotel guests in the PA for each year from 1999 through 2005, as reported by the Palestinian Authority.4 The intifada produced a quick, intense, and easily anticipated response: hotel stays—and tourism in general—plummeted. Estimates from the PA suggest a loss of 600 million tourist dollars between the second intifada’s inception in September 2000 and July 2002. Annual PA tourism revenue in that same period was only $300 million, so the loss was as large as the total tourist revenue. Keep these numbers in mind. We will return to them.
With these facts under our belts, we can work through the game-theory logic that points to the attractiveness of tourist tax dollars as a path toward peace. Imagine, for instance, that President Obama’s government or the United Nations presses the Israeli and Palestinian Authority governments to share with each other tax revenues arising exclusively from tourism and then administers the distribution of the funds. Each side’s share of those tax dollars is to be in direct proportion to its current proportion of the total Israeli and Palestinian populations in the PA and Israel.
FIG. 7.2. Tourism in the Palestinian Authority Since the 2000 Intifada
The tourist tax revenue arrangement need not last forever. It must include an irrevocable commitment for it to persist for a long time (say twenty years), and it is important that this tax revenue sharing be tied to a fixed formula based on the current populations and not on future changes in those populations. Opening the formula to renegotiation could create perverse incentives. It is also essential that the definition of tax revenue originating from spending by tourists be based on predetermined rules for estimating this source of income. Independent accounting firms might be used to provide a standard way to define and identify tourist revenue and the taxes derived from it. This tax revenue would then be allocated to each side over the agreed duration of the program based on a onetime fixed population-based formula with no questions asked.
Some tourist-derived tax revenues are obvious. Hotels check the passports of foreign visitors, and so, as we saw, it is straightforward to know how many tourists checked in, what their hotel bills were, and what were the tax portions of those bills. For every foreigner staying in a hotel—whether strictly a tourist or claiming some business purpose—one might stipulate that the taxes on that hotel stay go into the tourist tax revenue pot. It will be necessary to devise a good monitoring system to prevent underreporting, but that is probably something governments already have dealt with.
Restaurants don’t have as obvious a way of determining who their foreign guests are, but perhaps accountants can find a clever way to approximate the percentage of a restaurant’s taxes that is attributable to tourists. This might depend on the location of the restaurant, the proportion of foreign hotel guests in the area, the location of the bank for payments made by credit card, or many other criteria. The same might be true for shops. Those selling souvenirs, for instance, are likely to have more tourist-based revenue than those selling groceries. At passport control, visitors declare whether the purpose of their visit is business or pleasure. There, too, it is possible to create a revenue formula that approximates how much is spent by those saying they are tourists. Anyway, I am not trying to do the job of accountants, and I certainly am not qualified to do so. Accountants will be good at setting up sensible rules to identify the relevant sources of tax revenue, especially if their fees are also tied to that revenue.
On the population side, if Palestinians currently make up 40 percent of the population in the area, then 40
