The French Navy (FN)

11During the 1997-2002 period, the average technical readiness rate of the French Navy fleet lost more than 30 percentage points of readiness. Then, the readiness of the fleet increased from 2002 to 2006 but the deterioration in the readiness repeated again during the two years 2007 and 2008. Then, it increased to reach a maximum of about 78% in 2009. Finally, the current trend seems to be a decreasing one.

Figure 1

Average readiness rate of the French Navy (Excluding tanker-ships, Aviso and watching frigates)

Average readiness rate of the French Navy (Excluding tanker-ships, Aviso and watching frigates)

12However, these average rates mask both the worst and the best readiness situations. For example, the submarine SSN Class had readiness of only about 35% in 2001. In contrast, the readiness of transport ships seemed to be less impacted at the beginning of the 2000s.

The French Air Force (FAF)

13A decrease in the readiness is observed from 1998 to 2002 for fighter aircraft. The readiness of such aircraft increased from 2002 to 2004 and decreased again until 2008. In 2011, the average readiness rate of fighter aircraft was about 50%, however in a smoothly increasing trend since 2009. The readiness of support aircraft followed the same trend, however with an increasing gap that reached 10 points of readiness by 2011 (figure 2).

Figure 2

Average readiness rates of the French Air Force’s fleet (excluding Hawkeyes and training aircraft fleets)

Average readiness rates of the French Air Force’s fleet (excluding Hawkeyes and training aircraft fleets)

14The diversity (in age, use and technology) of fleets is a structural factor to consider in analyzing MRO processes and the technical readiness of the FAF. For example, the Air Force maintains very old tanker aircraft KC-135 (47 years), old transport aircraft like C-130 (30 years old) or C-160 (40 years old) along with the much newer Rafale aircraft (6 years old). Again, as we have already mentioned before, the global trends of the fleets we presented above should be combined with a full picture of the readiness of each fleet (or better each kind of aircraft). Finally, keeping in mind the previous observations, combat aircraft and helicopters were able to reach the targeted time of flight required by Air Force’s standards during the decade of 2000, which was not the case for transport aircraft (Dulait, Carrère, 2008).

The French Army (FA)

15In the 2000s, the French Army experienced a substantial drop in the readiness of its main systems. Light and heavy armored tanks experienced a global decline in readiness during this period (figure 3). Again, the “readiness crisis” is particularly noteworthy, especially if we look at the period 1997-2001. The data do not include the readiness of the heavy armored Leclerc tanks, whose rate was particularly low in the 2000s (under 40% in 2008).

Figure 3

Average readiness rates of the main armored vehicles in the French Army

Average readiness rates of the main armored vehicles in the French Army

16Another example is the Army helicopters whose readiness shortfalls were particularly extreme at the beginning of the 2000s (The army operates about 70% of the French State helicopters) (figure 4). This downward trend in the readiness of the Army helicopters coincided with a lack of practice for pilots, which compounded the problem.

Figure 4

Average readiness rates of the French Army helicopter’s fleet

Average readiness rates of the French Army helicopter’s fleet

The evolution of readiness as a consequence of a decrease in budgetary means?

17Readiness is related to budgetary means and the level of the budget credits gives a first look into the trends for resources allocated to the MRO process for the whole French Army. After an average increase of about 2% between 1977 and 1993, a decline of almost 20% on average is noted for the 1993-2002 period. Indeed, considering the trend at the end of the 1990s, the main cause of the decrease in the readiness of military platforms was first analyzed as a consequence of the decreasing trend in maintenance credits (see figure 5). Moreover, as figure 6 reveals, the decreasing trend in maintenance credits is more pronounced when compared to defense investment credits. It appears that, during this period, maintenance credits were often considered as an adjustment portion of defense expenditures for a defense budget under constraint. The consequences of these yearly trade-offs between maintenance credits and operating expenditure credits were called the “budget notch” in maintenance credits.

Figure 5

Maintenance credits trend (1977-2009, in million of Euros, AE-LFI)

Maintenance credits trend (1977-2009, in million of Euros, AE-LFI)

Figure 6

Ageing systems and obsolescence

24Most defense systems are now kept in service for many years. As the ageing of platforms is correlated to a rise in the number of breakdowns, the increase in the length of their lifecycle can be an important cause of the increasing cost of MRO. Figure 9 shows the heavy-maintenance workload ratio for several transport aircraft of the USAF. More precisely, the figure depicts the growth rate for a range of average fleet ages. The rate is expressed as a ratio of heavy-maintenance workload over time to the workload at the first heavy-maintenance inspection. In the FAF, some tanker aircraft are more than 45 years old, transport helicopters are about 30 years old and transport aircraft and Super Etendard fighter aircraft about 30 years. In the FA, light armored vehicles VAB are more than 25 years old on average with an increasing deployment and light gunship helicopters are about 28 years old. The FN maintains antisubmarine frigates that are about 30 years old and operates an ageing logistic ships fleet (e.g. oil tanker, repair ship). Ageing fleets impact significantly the cost of MRO.

Figure 9

Heavy-maintenance workload for large aircraft over time

Heavy-maintenance workload for large aircraft over time

Use and overuse of military systems

25The use for training and the increase of in operational missions are other reasons for the increasing costs of MRO. Since the first Gulf War, in rapid succession, foreign French interventions have particularly drawn on military platforms (mostly air and ground equipment). Such operations require a more intensive use of platforms and therefore accelerate their ageing. The latter phenomenon calls for more resources, raises the cost of MRO and finally impacts the readiness. Because of the involvement of France in multiple theatres, this former cyclical factor in the evolution of MRO cost has become a structural one.

Increasingly technology-intensive systems

26The modern defense systems are “systems of systems”. They typically integrate a complex combination of electronic systems, including defense, attack and security systems. Both procurement and maintenance costs of such systems are higher than the cost of simpler weapons. Complex “systems of systems” cannot be maintained using generic technologies and even less using generic tools. On the contrary, they need highly specific tools and infrastructure that are often very expensive. Therefore, the technology, in relation to the cost of inputs, influences the total cost of MRO (e.g. intermediate output, raw materials, spare parts and workforce). As a result, we assume the increasing technology and specific assets associated with its management to be a major cause of the increasing cost of MRO. Moreover, disruptive technology tends to impact the cost structure of the maintenance system in a lumpy, non-linear way. That is, costs evolve with big lumps in spending that occur all at once, rather than a little bit at a time or continuously.

On-board electronic systems

27In the case of modern and highly technological equipment, on-board electronic systems are an important driver of total MRO cost. Such electronic systems are more and more intensive in modern defense systems. For instance, the data processing capacity of the Rafale has no equivalent in current French defense systems. The maintenance of processing capacity in complex systems can be costly, especially in the diagnosis stage of the MRO process.

Composites

28Composite materials now represent a substantial portion of modern defense systems. Their development is driven by the need to enhance operational performance (e.g. through increased range, stealth, stability and payload). For instance, the share of parts that are made of composite elements has risen from 7% for a Mirage 2000 to 26% for the Rafale and from 2% for an F-15 to about 24% for the F-22 (Deo et al., 2001). The share of parts made of composite materials has also increased in naval systems.

29Composites tend to increase construction costs (e.g. about a range between 10-240% more expensive for a frigate superstructure (Mouritz et al., 2001). How their widespread use will affect MRO costs in the future is still an open question, however. On the one hand, composites theoretically reduce the ownership cost, for example by decreasing the number of MRO operations. Goubalt and Mayes (1996) found that the calculated life-cost of a composite boat are slightly less than about 7% compared to a steel boat of the same size. Through-life costs savings would seem to outweigh significantly any increase acquisition cost, but one has to keep in mind that composites also involve very specific and costly assets and might suffer from bottlenecks in supply considering the increasing demand for such elements in various manufacturing sectors (e.g. the civil aeronautical sector).

Workforce and skills related to the technology

30Third, workforce is an important element of the increasing cost of MRO. The cost attributed to the “workforce element” has two sides. The first one is qualitative and relates to the increasing complexity of MRO tasks and the skills necessary to deal with more and more complex military systems. Increasingly complex military systems need highly MRO skills that could hardly be found in the draftees or inside the Army. As a result, they need different skills very close to engineering skills, which are costly. In addition, the MRO time processes are likely correlated with complexity and cost. The second side is quantitative; with the professionalization of the Army, some basic but numerous tasks, formerly executed by the draftees (a workforce whose costs are largely not recognized in defense budgets) are now conducted by military or civilian professionals whose costs are clear.

Reforms and institutional changes

32Institutional processes, resulting from the establishment of the regular Army in 1997 or the transition from a public defense industry (i.e. arsenal organization) to a private one have mid-term consequences for the cost structure. They reveal the real cost of MRO; they may also increase it through “transition costs”. For instance, the French Army has achieved new visibility into its MRO costs due to the transition of GIAT, public industry, to the private firm Nexter Group (e.g. assessment of the spare parts, reorganization of the supply chain across national territory and social measures due to the downsizing of the workforce dedicated to military production). The French Navy experienced a similar situation with the transition from DCN to DCNS group. Such big changes call for a consideration of the “path dependency” in the management process of MRO (in logistics, skills, communication at work, spatial organization and history of facilities settlement, etc.).

Historical legacy

33First, due to the specific skills and assets associated with defense systems and their maintenance, only few firms are able to provide MRO services (prime contractors) or spare parts (sub contractors). In spite of the introduction of competition into requests for proposals for MRO processes, virtual monopolies still dominate the landscape, especially in the support of air platforms and in the naval sector. For instance, in the naval MRO, competition was the rule for 70% of the MRO contracts by volume; but by value, those contracts awarded under competition were estimated to be only about 20% of the total amount managed by the structure in charge of naval MRO. In 2011, DCNS group only had 30% of the number of contracts, yet they represented more than 70% of the total amount in euros dedicated to the naval MRO (Lamour, 2011).

34Second, as a legacy of an Army organization’s model developed for the Cold War, the location of military bases (mostly with their own MRO infrastructures) and unit installations (especially for the FA) were located in the northeast of France. In addition, spare parts were quite over-stocked and often located very close to the defense systems they supported. For 20 years, both the decrease in the size of the Army and its reorganization toward a “force projection” model have brought a change in the management of the MRO process, especially in its spatial side (e.g. warehouse locations, Army installations for staging equipment or goods to be shipped). As a result, some “military bases” have become more important than in the past for the French military system. To give a concrete example, the ports of Toulon and Marseille in the South of France have become the first force projection ports for the French Army, combining the Navy projection fleet with the proximity of the big Canjuers Army base. As a result, the former geographical patterns may not be in line with the current context and its economic and operational needs. Such inadequacy may yield costs mainly associated with distance.

35Finally, all of the factors we mentioned above impose constraints on the resources allocated to defense support. They may combine to create a complex, upward spiral of MRO costs, with consequences for readiness. For instance, the MRO process combines day-to-day maintenance of old equipment related to use and overuse, with obsolescence problems, in a context of increasing demand linked to the newest technology-driven systems. What we call the “New&Old problem” often involves the management of the MRO of two generations of defense systems in the same time. The increasing cost of spare parts related to the obsolescence, the cost of inputs per se and monopoly situations induce the maintenance structures to develop a kind of “platform cannibalism,” which in turn impacts readiness. Furthermore, among all the causes we identified in this paper, there is a kind of “glue”. More precisely, the organization of the MRO process is this “glue,” and it influences the cost-effectiveness. To improve this cost-effectiveness, the public planner can use various leverages that we present in the following section.

The “physical option”

37In the short run, one way to raise readiness rates obviously consists of reducing the volume of the supported fleets. The decrease in the fleets’ size automatically increases readiness rates, increases the number of spare parts (taken from former platforms) and allocates more resources to those systems that remain in service.

38We present the scope of the decreasing trend in volume between 1990 and 2010. During this period, the size of the FN workforce decreased by about 36% (from 72,136 to 45,986 personnel, with civil employees included and Gendarmerie excluded). Over time, the FN fleet has shrunk, from 110 ships in 1990 to 82 in 2002 and 70 in 2010. The number will probably drop to 60 by 2020.

39During the same period, the FAF decreased by about 39% (from 98,110 to 60,010 personnel, civil employees included, Gendarmerie excluded). In the same time, the number of fighter aircraft has been cut almost in half and the number of helicopters has decreased by about 20%. The number of transport aircraft remained constant between 1995 and 2010.

40Finally, between 1990 and 2010, the size of the FA has shrunk by about 56% (from about 329,234 to 144,490 personnel; civil employees included, Gendarmerie excluded). In the same time, the number of heavy tanks has decreased by about 80%. Light tanks have decreased by about 15% between 1989 and 2010. The size of the light armoured vehicle fleet fell between 1995 and 2010 from 7,000 units to about 5,200. Finally, the number of helicopters in the FA dropped from about 400 at the beginning of the 1990s to 312 in 2012.

41We argue that the volume effect certainly played a role in the improvement of the availability of systems. However, it does not address structural problems such as increasing costs or inadequate management systems, especially for technologically disruptive new platforms.

Budgetary means

42Another point of leverage for raising readiness is to increase the resources available for MRO. Returning to figure 6, notice that the ratio increased smoothly at the end of the 2000s (at a much later date after the first warming signs in the reports at the beginning of that period). However, the delay in reversing the trend likely magnified the financial effect in the “readiness crisis”. At the end of the 2000s, overall MRO costs (maintenance credits and wages) could roughly be estimated at about 15% of the French Defense budget (without retiree pensions) and at about 27% of investment expenditure.

43Figure 10 plots maintenance credit per active-duty military member. The current trend seems to be in support of increasing budgetary resources for MRO. For example, in 2009, MRO credits allocated to aeronautic support grew by about 10% compared to 2008 (Dulait, Carrère, 2009). That decision was made to cope with the increasing cost of highly technological systems such as the helicopter Tigre and its specific infrastructure in the FA. In addition, another report mentioned a doubling of maintenance credits between 2011 and 2012-2013 (Carrez, 2010, p. 56). But the simplistic rule that will connect more budgetary means to increasing readiness has never been modelled, and other more structural leverages have been set up.

Figure 10

Credit maintenance per “military capita” (excluding Gendarmerie – in thousands of euros – 1990-2015)

Credit maintenance per “military capita” (excluding Gendarmerie – in thousands of euros – 1990-2015)

New management models

44New structures and new management models also contribute to raise readiness and to reduce the cost of defense support. In recent years, the ministry of defense adopted a MRO based on fleets (Whole Fleet Management), with processes oriented toward the natural background of systems. To simplify, one deals with naval systems, another with air systems and the last with land systems. As a result, the French ministry of defense created specialized joint-army structures to optimize maintenance through “demand-oriented processes”. The establishment of new administrative bodies started in 2000 with the creation of the SSF (Navy Fleet Support Structure). At the same time, the SIMMAD (Structure of Integrated Aeronautics Management) was created to manage the aeronautic supply chain MRO for all three services. Finally, the French ministry of defense set up the SIMMT (Structure of Integrated Management of Ground Equipment) in 2010. These new administrative bodies permitted the ministry to gather all the skills and know-how under a single authority, with the possibility of developing new and potentially more efficient forms of management and partnerships.

45First, these new structures have adopted a lifecycle view involving all the actors of the “military supply chain”. Such an approach leads to the identification of potential cost items early in the weapon system design phase (e.g. reducing obsolescence risk through standardization methods) and therefore improves the potential for developing proactive and reactive initiatives.

46Second, such new administrative bodies also have developed “global contracts” gathering a pool of similar equipment (ships, aircraft, helicopters, tanks) under a single contract in place of multiple contracts. These new ways of managing the MRO supply chain seem to have led to a decrease in MRO costs, combined with an increase in readiness. For example, savings about 20% in the naval MRO with a significant increase of the readiness of the fleet were noted during the period between 2003 and 2008.

47Third, they also managed the “in service support” (ISS) of MRO. This support provides service “in theatre,” whatever the theatre is (real theatre or training theatre).

Economies of scale, size of defense facilities and “market power”

48Economies of scale are considered to be internal to productive units. They result from a decrease in the average production cost when production increases in volume. Such economies are often associated with the increasing size of productive units and with mass production (Fujita, Thisse, 2002). In the case of defense support activities, we identify three main origins for economies of scale.

49First, they are due to a better sharing of variable inputs and the distribution of fixed costs (e.g. overheads costs) across a larger base. In our case, some military infrastructures (administrative entities, warehouses, depots, maintenance specific infrastructures, etc.) are indivisible and their cost needs to be shared to lower the fixed cost per unit produced. Second, economies of scale can be improved through specialization of the workforce and the improvement in production processes due to both the division of labor and learning effects between workers. Third, we note economies that are not merely pure economies of scale, but consequences of larger size. An increase in the size of defense support facilities possibly concentrates buyers in a given area and gives them more market power and may help in price negotiation.

50Under the influence of economic forces (budget constraints, increasing cost of the support of defense systems and increasing cost in other defense support areas), some military sites have increased in size, in the search for economies of scale. For this reason, the spatial trend in the armed forces’ geography is one of concentration. In 2009, about 42% of the military was concentrated on 8% of the national territory. From 2009 to 2016, 83 French military sites have been or are expected to be closed and 33 other military structures are expected to move (on a total of 471 sites). Moreover, the military workforce tends to be concentrated in a few regions. Between 1990 and 1997, the number of Air Force Bases decreased by about 40%. Other bases closed in 2011, and 7 closures are planned in 2012. In the MRO fields, the infrastructures also tend to concentrate, especially for the most industrial levels of the MRO process, i.e., those with higher fixed costs. For instance, the MRO of the Rafale engines is made at the single site of Bordeaux in the southwest of France.

Agglomeration economies

51Agglomeration is the concentration of both producers and consumers in an area. In the search for localized increasing returns, agglomeration economies are considered to be external to economic agents. Following the literature, we assume the increasing returns to be a consequence of indivisibilities in the place where economic agents tend to agglomerate. The microeconomic reasons for agglomeration effects are complex and related to some mechanisms called “sharing, matching and learning” (Duranton, Puga, 2004). Agglomeration effects are related to pecuniary and technological externalities. Following the classical typology, which often divides agglomeration effects into “localization economies” and “urbanization economies”, we present agglomeration economies and connect them to defense support activities.

52“Localization economies” are external to a given economic agent but internal to the industry. First, since Marshall we know that externalities enhance the productivity of agglomerated agents, which benefit from each other due to their collocation. Collocated agents of an industry can exchange information, especially tacit information that can reduce coordination costs and ultimately raise individual productivity. They can also share specific inputs such as infrastructures or services that have high initial fixed costs. Moreover, agglomeration induces both improved specialization and adjustment of the workforce to local labor markets. Second, the increase in the size of the market (resulting from both the increasing size of a given production unit and the agglomeration of economic agents) makes it possible for producers to sell more intermediate or final inputs. Such localized (and non-tradable e.g. maintenance services), highly specialized and diversified inputs can lead to an increase of the overall productivity of the agglomeration (Fujita, Thisse, 2002).

53Third, the concentration of defense support activities in a “growing market” also may influence the location decisions of firms that wish to locate near prime contractors, or the expansion decisions of firms currently located near contractors. With such behavior, firms related to defense activities benefit from savings on both transport and coordination costs. For the regional economy, it might escalate from a “signaling effect” to a “snow-balling effect”.

54“Urbanization economies” are both external to economic agents and to the industry. In urban economics, such a phenomenon is related to the size of the city and often correlated to the density of agents. Such economies combine pecuniary and technological externalities. In our case, an increase in the size of military bases (and the market associated with their support) could lead to enhance agglomeration economies through both the concentration of specialized maintenance activities and the multiplication of links with other activities such as civil activities (e.g. educational and transport infrastructures). Concretely, with the possibly increasing size of cities related to bigger defense bases, the opportunities to share possibly common indivisible infrastructures to reduce fixed costs could become interesting.

55Finally, we call for going beyond the traditional concept of spinoff effects. The clusters resulting from the evolution of the traditional boundaries between the activities (e.g. civil & military boundary or public & private boundary), beyond the classical spinoff effects, could lead to more complex reciprocal spillover effects between military economies and their surrounding economies. Such a phenomenon is a conceptual innovation to analyze military and civilian interconnections and we call it “territorial duality”.

Limits related to spatial leverages

56First, economies of scale are under constraint of the volume to maintain. As for a given budget constraint, the increasing cost of technology is typically combined with a decrease in fleet size. That phenomenon induces less understanding of economies of scales. This is especially true for countries with small military fleets. One can compare the potential economies of scale of France in its nuclear aircraft carrier fleet (one unit) with those of USA (eleven units). Second, the search for economies of scale may favor the increasing size of defense support facilities that are concentrated geographically. But, the existence of transport costs between the shared indivisibility and dispersed military units will limit the potential for expanding their size. As a result, the “economic size” of defense support facilities could be determined through a trade off between economies of scale and transport costs.

57Moreover, as the defense organization is not merely the result of an economic process, the equilibrium between concentration and dispersion within the national territory should be analyzed with other non-economic forces (e.g. national doctrine for the coverage of the territory including operational and strategic considerations, local lobbying and hysteresis effects resulting from history). For example, though efficient from an economic point of view, the search for reducing the cost of defense support with economies of scale, but without considering operational or political issues, might lead to a non-optimal spatial pattern from a social perspective. Such a mechanism is all the more complex, in that the social sensibility to the geographic patterns of defense facilities depends on the collective assessment of threat, which always has been a difficult exercise in defense economics.