An investigation into the dynamics of certain tourism parameters in Mauritius.


Karim Jaufeerally, Institute for Environmental and Legal Studies, Moka, Mauritius

Abstract

The purpose of this paper is to investigate the link that exists between the number of bed places (BP) available in all hotels in Mauritius and the average number of tourists present on the island at any time during a given year (TOI) over the period 1986 - 2001. It is found that these two parameters are closely correlated with each other. It then becomes possible to estimate the number of additional beds that will be required with any increases in tourism arrivals. Furthermore data from the Mauritius Ministry of Lands and Housing shows the linear extent of coastal lands that were occupied by the hotel sector for a number of years. This data can be used to extrapolate the linear extent of coastal lands that may be needed in the future to accommodate any further increases in tourism arrivals. Hence it is found that for every additional 100,000 tourist arrivals an additional 8 kilometres of coastal lands is needed for new hotel construction. A new index is created of the average number of tourists on the island at any given time per bed place (TOI/BP). This new index is found to vary within relatively narrow limits over the time period considered (1986-2001). When a scatter graph of TOI/BP versus Bed Occupancy Rates (BOR) is made, it is found that the points are widely scattered over the graphical plane, but not in a random fashion. On a year by year basis, the data points appear very much to depict the trajectory of a dynamic system over time. An elementary mathematical analysis shows that the TOI/BP index is always greater or equal to the BOR. It is found that the TOI/BP index has declined over the last few years and is now very close to the historic average. In view of the planned expansion of new hotels in Bel Ombre for instance, the index may drop yet again unless accompanied by increases in arrivals. Such further drops in TOI/BP would signal a situation of over capacity in beds for the industry as a whole.

Introduction

Tourism is very important for the mauritian economy, in 2001 the country received 660,000 tourists, the industry generated Rs 18 billion in foreign exchange and employed nearly 20,000 people directly and indirectly (Central Statistical Office, 2001, Mauritius). Government and the private sector expect tourism to grow and contribute even more to the economy of this country. Indeed, currently (2003) there are more than a dozen hotel projects being seriously considered for Mauritius alone. Even Rodrigues has its share with another dozen being planned for this very small island of barely 100 square kilometres. Whether all these hotels will find sufficient tourists to maintain adequate bed occupancy rates is hardly debated locally as Government and the industry seriously expect close to 1,200,000 tourists annually by 2010. The social and environmental impacts of such an increase are assumed to be negligible compared to the anticipated benefits. There is a great need to research the issue of tourism dynamics and its environmental impacts for obvious reasons, the island being so small, any severe environmental or social degradation caused by tourism could impact upon it rather quickly. Hence it is critical to avoid overshoot of this industry.

Method 

The data used for this paper is readily available from the Handbook of Statistical Data on Tourism 2001 published by the Ministry for Tourism and the Central Statistical Office. The data used are (1) Total Tourists Nights, (2) Bed Places Available in all hotels from 1979 to 2001, (3) bed occupancy rates from 1986 to 2001. The data on coastal lands under occupation is available from the Ministry of Lands and Housing.

The first variable to be calculated is the average number of tourists on the island at any one time during a given year (TOI). This is calculated by taking the total number of tourist nights spent during a given year and dividing that by 365 which is the number of nights in a year. In doing so, one in effect gets the average number of tourist nights spend each night during the year which is exactly equal to the average number of tourists on the island during the same year. This variable is plotted against bed places (BP) available. A very high correlation is noted to exist between these two variables.

The TOI variable is then divided by the number of bed places available during the same year. In doing so, the average number of tourists per bed place available is found. This new index is called TOI/BP and is calculated for the years 1979 to 2001. This data can be plotted as a time series from 1979 to 2001. As will be discussed later on, interesting trends emerge from such an analysis. Finally, TOI/BP is plotted against the Bed Occupancy Rates (BOR) for the period 1986 to 2001 (BOR data is unobtainable for years previous to 1986).

 

Results and Discussions

Over the years, TOI increases from 3,889 in 1979 to 17,884 in 2001. A four-fold increase, in line with the stupendous increase in tourism arrivals which shot from 128,360 to 660,318 over the same period. As can be expected the number of bed places increases from 3,888 to 18,350 for the same period. When BP is plotted against TOI (figure 1), it is found that they are highly correlated (R2 = 0.9758, where R2 is the coefficient of variation). A perfectly reasonable finding, the more tourists there are on the island, the more bed places there will be to accommodate them. What is surprising is the very high value of the coefficient of variation so calculated (0.9758). A straight line equation can be fitted to the data linking TOI to BP by regression analysis. That equation is TOI=1.0167 x BP - 264.14 (eq 1). Over the last few years, on average, each tourist spent 10 nights in Mauritius. Assuming regular increases in tourism arrivals, the number of beds required to cater for such increases can be readily calculated. For example, should arrivals reach 700,000 tourists in a given year, with an average of 10 nights per tourist that would make 7,000,000 nights spent in Mauritius during the year. The average number of tourists on the island at any one moment during the year then would be 19,178. This number of tourists on the island would require 19,123 beds according to eq. 1.

 

 

 

This calculation can be repeated for greater number of yearly arrivals as shown below:

Future Tourism Arrivals

Future TOI (assuming 10 nights per tourist) TOI= (Arrivals x 10)/365

Bed Places required from eq 1.

700,000

19,178

19,123

800,000

21,918

21,817

900,000

24,657

24,512

1,000,000

27,397

27,207

1,100,000

30,137

29,602 

1,200,000

32,877

32,596

 

Now from the Ministry of Housing and Lands it is possible to know the length of coastline under hotel occupation for a number of years as is shown in the table below.

Year

Bed Places in all Hotels

Tourism Arrivals

Length of Coastline under hotel occupation (Km) , Source: Ministry of Housing and Lands

No of Beds per Km

1990

9572

291,550

29

330

1996

13,833

486,867

41.9

330

1998

14,995

558,195

44.5

337

Average

 

 

 

332

 

Using the above data, the number of hotel beds per kilometre of coastline under hotel occupation is readily calculated. It varies from 330 beds per km to 337 beds per km with an average of 332. It is interesting to note how the number of beds per kilometre has varied so little over the nineties. It must be pointed out that not all hotels are directly onto the beach head, there are a number of off beach hotels along the coastline and a number of inland hotels too, in the capital city Port Louis for instance and elsewhere. The number of beds per kilometre must therefore be taken as a global statistic that links bed places in all hotels and the length of coastline under hotel occupation.

Using the average number of beds per km, it is possible to make a reasonable estimate of the length of coastline which will have to be devoted to hotels for future arrivals. This assumes that the mix of beach hotels and off beach hotels stays roughly the same in the future. The total linear extent of the Mauritian coastline is estimated to be 323 kilometres (Ministry of Lands and Housing), the percentage of coastline which will have to be devoted to hotels is also calculated.

Future Tourism Arrivals

Future TOI (assuming 10 nights per tourist) TOI= (Arrivals x 10)/365

Bed Places required from eq 1.

Length of Coast line required for hotels assuming 332 beds per km

Percentage of Mauritian Coastline devoted to hotels

700,000

19,178

19,123

57.60 

17.8 %

800,000

21,918

21,817 

65.70 

20 %

900,000

24,657

24,512 

73.80 

22.8%

1,000,000

27,397

27,207

81.95 

25.4%

1,100,000

30,137

29,602 

89.15 

27.6%

1,200,000

32,877

32,596

98.20 

30.4%

 

From the above figures it is clear that any increase in arrivals will be associated with an increase in coastline under hotel occupation. Indeed, from the above figures it can be inferred that approximately 8 kilometres of coastline are needed for hotels for each 100,000 arrivals. Considering that the Mauritian coastline is not that extensive (323 km) in length and that not all of the coastline is made up of high quality sand beaches, it is obvious that further hotel construction will compete increasingly with the public's need for better access to beaches. This in turn will lead to more and more local opposition to further hotel construction. Indeed, this is already the case and has been the case for the past ten years, where most new hotel construction directly onto the beach head has met and meets with increasing local opposition though it is largely ineffectual as nearly every hotel project has gone through. Further conflicts between the hotel industry and the general public may slowly erode the goodwill of the latter towards tourists who may then be perceived as invasive. A state of affairs that is not desirable by all parties concerned.

Further information can be extracted from TOI data. TOI can be divided by BP (number of bed places) for each corresponding year from 1979 to 2001. This new index, the TOI/BP index varies from a low 0.786 in 1983 to a high 1.188 in 1989, with a historic average of 0.973. This ratio or index measures the number of tourists there is on the island per bed place in the hotel industry. It is a measure of the potential demand for hotel beds because when the index is low, there are few tourists around for the beds available and so one cannot expect a large demand for hotel beds. However when it is high there are many more tourists around for the beds available, consequently one can expect a high demand for hotel beds. In effect the index is a measure of potential demand for hotel beds by tourists in Mauritius. When this index is plotted against time, interesting trends emerge, figure 2. It begins in 1979 with a value of 1.00 and then drops rapidly to a low 0.786 in 1983. It then climbs up quite rapidly to a high 1.188 in 1989, then oscillates back and forth with each peak successively lower than the previous one and in 2001 it is 0.975, very close to the historic average. Over all, this index displays a waveform reminiscent of a damped oscillation that with time seems to settle close to the historic average.

From 1979 to 1983, arrivals fell from 128,360 to 123,820, whilst bed places increased from 3888 to 4900. This explains why the index fell steadily. However as from 1984 onwards arrivals increased year by year, therefore increasing TOI. The number of bed places rose also but not as fast as TOI because presumably it takes a few years to respond to any increases in arrivals. Therefore a steady increase in TOI/BP. After a few years the demand was satisfied and so the TOI/BP index fell. Further increases in tourist arrivals made the index rise but further hotel construction placed even more beds on the market, making the index fall back and so on. The fact that the index is now close to the average and that each successive peak was lower than the previous one could indicate that demand is being satisfied and that further increases in available beds could depress the index below the historic average. A depressed index could signal that a situation of over capacity has been reached. Although the official figures for 2002 are still not out, provisional figures are as follows: arrivals: 681,648, visitors nights spent: 6,768,870, beds available: 19,597, bed occupancy rates: 0.59. This computes as a TOI/BP of 0.946 for 2002, which is below the historic average. If over the next few years the number of beds available increases as planned and if arrivals do not increase in steps, the TOI/BP index will fall even more and the hotel industry will find itself in an over capacity situation.

When the TOI/BP index is plotted against bed occupancy rates (BOR) for 1986 to 2001 (previous to 1986 there are no official BOR figures), a scatter graph is obtained and although no definite trends are obvious, there is a distinct pattern to it (figure 3). Linking each point year by year, the curve depicted resembles the trajectory of a dynamic system. Earlier, it was argued that TOI/BP is a measure of potential demand for bed places, when it is above average, the potential demand for bed places is high, when it is below average, the potential demand is low. On the other hand, the Bed Occupancy Rate (BOR) is very much a measure of supply of bed places. When the BOR is above average, hotel beds are in short supply, when it is below average, hotel beds are in plentiful supply. Hence the scatter graph is very much a diagram that depicts the potential demand and supply of hotel beds in Mauritius and how it evolves over time. The graphical plane can then be divided up into four quarters along the historic averages of TOI/BP and BOR. In effect the following combinations are as follows: (1) Low TOI/BP, Low BOR, (2) Low TOI/BP, high BOR, (3) High TOI/BP, Low BOR, (4) High TOI/BP, High BOR.

It is interesting to note that TOI/BP is always greater or equal to BOR as shown below. By definition visitors nights (VN) spent during a given year is always greater than or equal to hotel nights (HN) spent during the same year.

 

 

By definition

VN >= HN (eq 1)

Now VN / 365 = TOI

So

VN/365 = TOI >= HN/365 (eq 2)

Now dividing both sides of eq 2 by the number of bed places (BP) available in all hotels during the same year,

TOI/BP >= (HN/365) / BP

Now (HN/365)/BP = BOR

Therefore TOI/BP >= BOR (eq 3)

This is a physical constraint of the situation that can never be breached.

A closer analysis of the scatter graph and of the four quarters reveals a number of interesting features. Let us examine the first quarter where both TOI/BP and BOR are below their respective averages. A low TOI/BP would imply that potentially there is a low demand for beds as there are fewer tourists around per bed. A low BOR would imply that the supply in beds is above average as there are lots of beds left unoccupied. Such a state of affairs would be found in a stagnant or declining industry.

The second quarter is when TOI/BP is low and BOR high (meaning lower and higher than the historic averages respectively), meaning that the demand for beds is low, but yet beds are in short supply. It sounds contradictory and most probably is to a certain extent. But it could simply mean that though there are few tourists per bed, the hotel sector manages to attract most of them in its establishments. In effect, there is a swing towards hotel accommodation.

The third quarter is when TOI/BP is high whilst BOR is low. This could happen when there are lots of tourists per bed but they choose to remain in non-hotel accommodation for a number of reasons, such as high prices for hotel rooms and / or low overall standards in the hotel industry. The swing is away from hotel accommodation.

The fourth quarter is when both TOI/BP and BOR are high. A high TOI/BP means that the demand is high for hotel beds and a high BOR means that most beds are sold out and so there is a short supply of beds. Obviously this is the ideal situation. Hotels are full, lots of tourists around, high prices and therefore a fantastic stimulus for further growth in hotels.

 

 

 

 

The following table summarises the above discussion.

Table 1

 

 

TOI/BP

TOI/BP

 

 

LOW

HIGH

BOR

HIGH

Swing Towards Hotel Accommodation

Growing Tourism Industry

BOR

LOW

Stagnant or declining Tourism Industry

Swing Away From Hotel Accommodation

 

The above discussion shows that the scatter graph of TOI/BP versus BOR depicts very much the state of the tourism industry and the different phases it is in. It is a phase diagram of a dynamical system. If we examine more closely the phase diagram for the period 1986-2001, we note that most data points are in the high TOI/BP and high BOR quarter, with very few points in the other quarters as table 2 shows. This indicates that the tourism industry for this period was very much in a growth period. Indeed this is exactly what happened in Mauritius during that period. The last two decades have seen spectacular increases in tourism arrivals and in hotel beds available. It was a period of tremendous growth and opportunity. This concordance between theoretical considerations of the meaning of the four quarters and the actual data points is encouraging. It implies that the theoretical analysis is not completely off the mark.

Table 2

 

 

TOI/BP

TOI/BP

 

 

LOW

HIGH

BOR

HIGH

1

9

BOR

LOW

2

4

 

For 2002, the provisional figures are TOI/BP, 0.946 and BOR, 0.59. This places the 2002 data in the first quarter, signifying a drop in potential demand yet the hotel industry managing to retain a bit more tourists. This implies that a swing towards hotel accommodation has occurred. Over the next five years, up to a dozen new hotels are expected to open up, eight in Bel Ombre (in the south of the island), two to three on the west coast and a few more in Rodrigues. This should put on the market between 4000 to 5000 beds, increasing beds available by 20% to 25%. It is obvious that TOI must also increase by a similar quantum otherwise the TOI/BP index will fall significantly below the historic average. In fact to return to the high TOI/BP and high BOR quarter characteristic of the growth years, TOI must increase faster than BP. Assuming that from 2001 to 2006, the available beds will increase by 20% and assuming that the TOI/BP index should climb to 1.00, a little above the average, then TOI must increase by 23%.

Such an increase implies a compounded yearly increase of at least 4%. It might be difficult to achieve considering the present difficult world economic situation and the fact that over the last five years the yearly percentage increases in TOI have been between 3.69% to 1.79% except for 2000 where the increase was a massive 11.93% as shown in table 3. Failure to achieve such growth will surely mean that the industry will find itself with an excess of beds.

Year

TOI

% increase in TOI

1997

14935

 

1998

15254

2.14

1999

15697

2.90

2000

17570

11.93

2001

17884

1.79

2002

18545 (provisional)

3.69

 

 

Conclusion

In this paper it was shown that there is a very high correlation between the number of bed places available in hotels in Mauritius and the average number of tourists on the island at any one time during a given year. Hence, it becomes possible to calculate the number of bed places that will be required for any future increases in arrivals. It was also shown that for every 100,000 arrivals, 8 kilometres of beach frontage is used up for further hotel construction. Given the limited size of the island, future conflicts between public and hotel promoters are inevitable and that will be an ongoing feature of this industry. It was also argued that the TOI/BP index is a measure of potential demand for hotel beds. Whereas BOR can be considered as a measure of supply of beds available. When plotted against each other the resulting scatter graph gives rise to a phase diagram of potential demand versus supply. It traces the evolution of the tourism industry with time. Using this phase diagram, it can be argued that the industry as a whole is close to over capacity. Should planned increases in hotel construction not be matched with increases in arrivals, the system may enter a stagnation phase from which it might be difficult to climb out of. The next few years will be critical for this industry as a whole.

 

 

 

References

(1) Central Statistical Office and Ministry of Tourism, Handbook of Statistical Data on Tourism, 2001, 2000, 1999, 1998

(2) Central Statistical Office, International Travel and Tourism Statistics, 1998, 1996, 1995, 1994, 1993, 1992, 1991, 1990

Appendix

Year

Tourist Arrival

Visitor night spent

Bed Places (BP)

BOR

TOI

TOI/BP

1979

128,360

1,419,610

3888

3,889

1.0003

1980

115,080

1,301,730

4000

3,566

0.8916

1981

121,620

1,361,190

4484

3,729

0.8317

1982

118,360

1,392,510

4530

3,815

0.8422

1983

123,820

1,405,870

4900

3,852

0.7861

1984

139,670

1,541,590

5102

4,224

0.8278

1985

148,860

1,735,960

5387

4,756

0.8829

1986

165,310

1,878,370

5955

48

5,146

0.8642

1987

207,570

2,371,970

6418

54.8

6,499

1.0126

1988

239,300

3,002,620

7005

58.2

8,226

1.1744

1989

262,790

3,196,780

7374

58.3

8,758

1.1877

1990

291,550

3,564,930

9572

54.3

9,767

1.0204

1991

300,670

3,696,990

10482

47

10,129

0.9663

1992

335,400

4,110,430

10917

47

11,261

1.0316

1993

374,630

4,610,400

11058

53

12,631

1.1423

1994

400,526

4,359,303

12187

60

11,943

0.9800

1995

422,463

4,434,891

12359

59

12,150

0.9831

1996

486,867

4,957,683

13833

60

13,583

0.9819

1997

536,125

5,451,314

14126

64

14,935

1.0573

1998

558,195

5,567,889

14995

63

15,254

1.0173

1999

578,085

5,729,464

16947

62

15,697

0.9263

2000

656,453

6,412,876

17776

62

17,570

0.9884

2001

660,318

6,527,800

18350

58

17,884

0.9746

2002 (prov. figures)

681,648

6,768,870

19597

59

18,545

0.9463