Demographic Analysis of the Makkah Province for the Purpose of Evaluating the Balance of the Urban System

In this paper, it is not feasible to make the review of a massive body of research, investigations and tests on the hierarchical distribution of cities and their importance in complex urban systems. So instead, we will try to introduce the most famous researchers who have initiated this track of studies and cite some recent cases of theoretical scrutiny and practical application.

As regards primacy, Mark Jefferson (1939) first defined primate cities of a different order of magnitude and the significance of other towns in a national urban hierarchy. As an example, he investigated the case of London (1931), which had a population seven times larger than the second city in Britain, then Vienna (1934), which had 12 times the people of the second city in Austria and finally Copenhagen (1935), which had nine times the population of the second city in Denmark.

Since the classic work of the German geographer Walter Christaller (1933), who introduced the central place theory, there has been an increasing interest in the size and frequency distribution of centres in urban systems. In addition, numerous empirical studies have displayed evidence of a functional hierarchy.

For the hierarchal distribution of cities in urban systems, Zipf’s rank-size law (1949) is one of the most striking empirical laws in urban economics, describing a remarkably stable regularity in the spatial structure of market economies (Krugman 1996). His studies in the United States imply a unique rank-size relationship according to which the area’s largest city (New York) is twice as large as the second-largest city (Los Angeles), three times as large as the third-largest city (Chicago), and so on. On the other hand, Beckmann’s work was the first attempt to estimate city size hierarchies in the central place theory. Moreover, Beckmann’s analysis is the first initiative to introduce the rank-size law in the existing spatial economic ideas. In the 1950s, the works of Beckmann explore the emergence of rank-size law from the ‘Lösch’ construction to the recent comparisons of ranking coefficients of different distributions. The latter allows the description of the hierarchical organisation of systems of cities. Parr (1969) was also an early pioneer of the theoretical discussion of city size hierarchy.

These theories of hierarchal regularities in urban systems have also been the focus of many critical recent contributions. For instance, Pumain (2004) explained the hierarchical fact by the existence of several levels of service defined by spatial ranges (a distance that the consumer agrees to travel to obtain the service, determined by the additional cost of transport added to the purchase of the product.

The theoretical explanation and practical testing of these rules, especially that of Zipf, have also come to life again, theoretically and empirically. The principal theoretical insight has been the understanding of the extent to which the rank–size rule is simply a statistical phenomenon with virtually no empirical (or economic, or regional science) content. In particular, Reed (2002) and Gabaix (1999) have suggested that the rank-size rule is the natural result of a growth process that is independent of the size of the city. Reed (2002) has also found that the rank-size relationship could vary with different signs, depending on which tail of the distribution one is examining. Fujita (1999) has argued that the rank-size rule approximates the long-run spatial distribution of a mature spatial system. Giesen and Suedekum (2009) have claimed that it is not only appropriate for Germany’s national urban hierarchy, but also for the city size distributions in single German regions.

Yanguang (2010) has integrated Zipf’s law to allometric scaling and fractal relations framework based on a hierarchy of cities in China, and then proposed a model of playing cards to explain the Zipf distribution and hierarchical scaling.

Jefferson (1939) has identified in his later research the conditions under which Urban Primacy, the direct consequence of an unbalanced hierarchy in urban systems, might occur.

Many other researchers have also conducted investigations into the factors underlying these urban irregularities in many cities worldwide. For instance, Mehta (1964) found that urban primacy resulted from the small area and population size of countries, while Linsky (1965) argued that high urban primacy was most often correlated to a small extent but high density, low per capita income, export-oriented and agricultural economies, a colonial history, and rapid population growth rates (Short, Pinet-Peralta 2009). Ades and Glaeser (1995) explained urban primacy by political factors such as instability and high levels of centralised political power. Moomaw and Alwosabi (2004), who restricted their investigations to Asian and American cities, showed that primacy was associated with economics and population size, economic development, population density, level of industrialisation and capital city status. Krugman’s (1996) general model suggests that primacy decreases with the openness of a national economy, but this has not been supported in the literature.

In our current investigation, we refer to both Zipf’s and Beckmann’s theories as they are the most common methods used globally to investigate the hierarchical balance in complex urban systems.

When talking about Makkah, it is important to emphasise two different urban scales attributed to this name. The first is related to the ‘region’ or the province of Makkah, and the second is associated with the constituency of the city (Fig. 2), where the Haram and the Holy Mosque are located. It is indeed the ‘Emirate’, as the capital of the region.

Fig. 2

The region or the Makkah Province is made up of 17 governorates, including that of Makkah (Fig. 2). The city of Makkah is its administrative capital. Each governorate is assigned centres of different categories (A or B) based on the level of development and the number of facilities they host. This categorisation is also directly linked to the urban area of the centre, the importance of the population residing there and the urban services required. The population of different governorates of the Makkah Province is distributed as shown in Fig. 3.

Fig. 3

According to the General Authority for Statistics (GaStat), the population in the region of Makkah includes Saudi citizens representing 59.52% of the total, with the rest being multinational foreign residents with a percentage of 40.48%; their distribution is given in Table 1.

In the governorate of Makkah and since the building of the House of Allah ‘the Kaaba’ by the prophet Abraham, the inhabitants settled all around it because of its spiritual dimension and the sacred religious connotation which has always been attributed to this building, as it is considered one of the most attractive places in the holy city, Saudi Arabia and the whole Muslim world (Fig. 4). With the spread of Islam around the world and the development of transport, the importance of the Kaaba increased. It became the destination of millions of worshippers from all over the world. The urban areas continued to be dense, leading to an ultimate state of urban overload compared to the neighbouring urban centres.

Fig. 4

This urban expansion in Makkah city is accelerating over the years with a very ascending frequency (Table 2).

Only 3 centres among the 131 that make up the region are at this level. These are the primate cities of Jeddah with m = 1, Makkah with m = 1.27, and Al-Taef with m = 1.49 (Fig. 6). In this level, the curve shape is nearly horizontal since the values of the ms are relatively close to the unit. Although the values of the constants of the two second cities are <1, it is essential to emphasise the relative coherence between the elements of this group of centres. However, a global reading of the values and the curve reflects a macrocephaly of the urban system.

Fig. 6

Compared with this level, it is noted that the values of m deviate considerably from the value of the unit balance. The striking inconsistency means that the slope of the curve takes a very irregular shape compared to that of the previous level (Fig. 7). There is a relatively remarkable demographic deficit relevant to the primate city of Jeddah and the city of Makkah. This level fairly represents the entire system in that it groups the highest number of urban centres in the region (82 centres out of 131). This group displays an unbalance compared to a normal distribution of the population in the system under study.

Fig. 7

There are four centres with a demographic constant between 100 and 170 (Al-Zarban, Al-Michaan, Al-Sahm, and Mashash al-Taref) (Fig. 8). This value of demographic constant shows a critical shortage in the population. However, there is a relative consistency in the shape of the curve at this level, which is nearly horizontal. The curve seems to reach its peak, which corresponds to the most unpopulated centres that require a proactive correction to achieve as quickly as possible the demographic size corresponding to both their ranks in the demographic hierarchy and their coherence compared to the size of the primate city.

Fig. 8

After the relative consistency observed in the previous level, it is noted that there is a sharp rise towards another level in only two centres, whose demographic constants are >1,000. The difference between the unit referential value of the primate city and the two last units of Al-Khadiraa and Warcha, which have demographic constants equivalent to 1,601 and 2,516, respectively, shows a chronic deficit in the urban system (Fig. 9).

Fig. 9

Unlike Beckmann’s theory, Zipf’s law does not consider the primate city as a reference unit for the whole system. “The primate city cannot be taken as a reference since Zipf’s law is rather a law of internal and hierarchical distribution within the urban system (Belhedi 1992).

The population of the primate city is undoubtedly an essential element of the system, but it cannot be a basis for the distribution effect. On the contrary, considering the total population as a reference base, the first city would be determined like the other cities. Zipf’s law highlights a simple relationship between the population of urban centres of the same network and their hierarchical ranks (law/rank-size). It is essential to remember that the law is often presented in its simplified form: Pr = P1/r, where P = population of the primate city, n = urban centre and Pr = Population of centre n. In reality, this model defines an ideal hierarchical distribution that a balanced urban system should have. “The regularity is in a way that the second city is half the first and the third city is only a third thereof. The size of cities in a given country follows an arithmetic series: 1, 1/2, 1/3, 1/4¼ 1/n”(Belhadi 1992). To apply Zipf’s law to the region of Makkah, we modelled a bi-logarithmic orthonormal marker that allows covering the large variations and ranges of values in the abscissa, referring to the ranks of urban centres and the ordinate and their populations. The curve obtained displays the actual distribution of the urban centres of the Makkah region based on their growing demographic hierarchy. To evaluate the result based on this model, we have inclusively drawn the adjustment line. This line passes through the majority of the projected points in the curve. It represents the look that the curve should theoretically have, according to Zipf. Comparing the adjustment line with the rank-size curve allows us to read the differences between Zipf’s theory and the reality on the ground (Fig. 10).

Fig. 10

At the top of the curve, three centres are isolated from the rest of the components of the Makkah system. These centres are the cities of Jeddah, Makkah and Al-Taef, which occupy the first three hierarchical ranks and have a population of 3,955,839; 1,547,843 and 883,870 inhabitants, respectively. Their positions in the orthogonal bi-logarithmic coordinate system are largely below the adjustment line, which confirms not only their accentuated demographic plethora but also their hypertrophy. This hypertrophy is justified by the fact that all the rest of the human settlements in the Makkah provincial system are below the adjustment line and are therefore in deficit except for the centres of Group 3, which seem more or less coherent since they are projected on the adjustment line, or they experience a very slight plethora for some centres. The convexity of the first group of centres and the projection of their positions in the ordinate axis makes it possible to precisely determine their plethoric values (Table 4).

Therefore, the first three cities are hypertrophied. However, it is necessary to highlight the importance of the plethora, which exceeds 1,000,000 inhabitants for Makkah and 3,000,000 inhabitants for Jeddah.

In this group, it is possible to note the difference in demographic sizes between the first two cities, which significantly exceeds the limit stipulated by the Zipf model in which, theoretically and in a balanced hierarchical distribution, the ratio between the sizes of the first and second city of the system should barely exceed 2.

However, the interpretation of the obtained results shows that this ratio is equivalent to 2.55. This indicates that there is not only another internal hypertrophy in Group 1 but also a remarkable macrocephaly of the urban system of the Makkah Province due to the demographic plethora of its primate city compared to the other successive urban units.

Generally, the centres in Group 2 have smaller demographic sizes than those required by their ranks in the demographic hierarchy. Starting from the fourth rank in the system, we note a move from plethora to deficit. The difference between the first three major cities and the fourth city is more significant than that which results from the application of Zipf’s law. Looking at the abrupt change in the curve between the first and the second group, we assume that this is related to an apparent deficiency in urban areas with populations between 100,000 and 800,000 inhabitants. The steepness of the slope and the absence of centres in this hierarchical level translate to the theoretical vacuum, which is proof of the inadequate demographic distribution at the top of the urban system (Fig. 11).

Fig. 11

We argue that the demographic growth can be explained by the intense competition between the centres with a significant urban presence (Jeddah and Makkah). This competition increasingly diminishes the population of other urban areas that have significantly smaller sizes. Therefore, they do not allow them to reach a demographic threshold that agrees with the theory, which takes into account a uniform space with a rational hierarchy of urban units based on their demographic weights. For example, the fourth-largest city, Bahra, should have a demographic size equivalent to a quarter of the population of the primate city of Jeddah to achieve a regular and consistent demographic size.

By applying the law Pn = P1 / Rn, this means that P4 = 988,959, whereas the actual value of the number of inhabitants in Bahra, fourth in rank at the regional level, is equal to 80,307. The difference is indicative of the hiatus which separates the first two groups. On the other hand, this city occupies a strategic geographic location because it is very close to metropolitan centres with an assertive economic dynamism. It constitutes a central axis of communication between the two major regional metropolitan cities, namely Jeddah and Makkah. The importance of this centre is precisely overshadowed by the juxtaposition to the primate cities that dominate it. It seems that the cities of both immediately higher and lower ranks had probably limited their growth. The city of Jeddah is growing due to its strategic geographic location, and the city of Makkah is growing due to its religious connotation.

This level groups the centres which are relatively balanced according to Zipf’s law with some overall non-significant gaps. Their populations count between 4,000 and 40,000 inhabitants. Compared to their ranks, they have demographic sizes which are consistent with the requirements of Zipf’s law. Although the number of centres in this group represents 42% of the total number of urban units in the system, the percentage of settled populations represents only 9.6% of the total population of the province.

This level, in which the units clearly show a deficit in terms of the population, includes 47% of the total number of urban centres in the system. This is the largest percentage of all the groups. Given the quantitative importance of the centres in this group, the sum of the demographic sizes of all the centres that compose it does not exceed 1.4% of the total population of the system. The most distressing report which indicates a striking inequality in the distribution of the population and an alarming disproportion between the top and the bottom of the curve is undoubtedly that of the last two centres in the demographic classification, namely Warcha and Al-Khadiraa: these two centres are isolated from the rest of the urban units, including those suffering from underpopulation and belonging to the immediately higher ranks. Regarding the rank-size model used in this analysis, the difference between the two centres and the rest of the components of the urban system is huge. At this level, the imbalance is amplified and the macrocephaly of the provincial system is accentuated.

A comparison of the real and theoretical distributions (Fig. 13) shows that the Makkah Province has several weaknesses regarding the distribution of demography over the different urban centres that form it. The variations in the curve show a significant imbalance compared with the theoretical adjustment line. Compared to the primate city, the second city is not developed enough to attract a substantial number of the population. It has only 1.5 million inhabitants, a deficit of around 500,000 for its rank as the second city after Jeddah. It also shows the non-existence of the third and fourth cities in the current distribution of the city system in the Makkah Province. This non-existence reflects a hierarchical hiatus and the lack of intermediate centres that meet the Zipf model requirements. At the base of the pyramid, the demographic weight and the number of centres per level are insignificant and far from the values stipulated by the theory (Fig. 12).

Fig. 12

Fig. 13

The demographic growth of the primate cities of the urban system of the Makkah Province is closely linked to the religious importance that the holy city has, being the first sacred city of Islam. The sacredness of the ritual of ‘Hajj’ and ‘Umrah’ makes it an ultimate destination for faithful Muslims. These rituals attract not only residents of all of Saudi Arabia, who visit the city more frequently and easily, but also visitors from elsewhere whose number, despite the fluctuations, remains considerably significant. The political vision adopted recently by the sovereigns of the Kingdom facilitates access to the holy places and therefore, it amplifies the demographic impetus of the primate cities and consequently the macrocephaly of the whole province previously demonstrated by the two used models of Zipf and Beckmann.

After the rapid economic development of Saudi Arabia following the discovery of petroleum, the forts surrounding the urban centres have been destroyed. “The fort of Jeddah was destroyed in 1947 and that of Riyadh in 1950. After that, other operations of the destruction of the fortifications which surrounded the cities and the villages took place because of the sudden and accelerated urbanization” (Al Abyadh 2016). This has contributed to the densification of cities and their transformation into large metropolitan areas. Then, there was an unprecedented migratory flow from rural areas to big cities, mainly Riyadh, Makkah and Jeddah (primate city hypertrophied in this urban system). This rural exodus was accompanied by the external immigration of hajjis and the faithful Muslims, a large part of whom settled in the capital Jeddah because of its proximity to Makkah and for the quality of services that it offers to its residents. These migratory movements give Jeddah an excessive urban image compared to the rest of the centres, which has contributed to making the province a macrocephalus system.

“Arabia, in the absence of rainfall and fertile agricultural lands, has never been a society the economy of which is primarily based on agricultural activity. Being a harbor city par excellence, the Hejaz region (Jeddah and its region) is much better known for its commercial activities” (Ladraa 2016).

In the Makkah Province, commercial activities are of utmost significance. According to the Saudi GaStat (2017), Jeddah city, the economic and tourism capital of the Kingdom holds nearly half the commercial units of the entire province with 50,365 businesses. Many factors have underlined this pre-eminence, for instance, the wide range of facilities and equipment including transport. This has given Jeddah great influence and radiance over the rest of the urban centres, including that of the religious capital ‘Makkah’ and enhances tremendously its demographic primacy.

Although theoretically, there is a division of the territory into administrative districts where representatives of the state are appointed in the various centres of the province, we find that this division does not correspond to an efficient political will that provides these urban units with autonomy. The political system gives the central government the power to decide. Although there are equivalents of local governments, in the context of Saudi politics, they are not sovereign and are dependent on the central government. This dependence is often the result of a shortage of local capacity and the lack of resources necessary to apply a self-governing policy. The Makkah Province, like other provinces of Saudi Arabia, remains centralised despite its evolution towards territorialisation. This centralisation affirms this context so that it favours the monopolisation of capitals and, therefore, urban macrocephaly.