Impact of Financial Flows: Emergence of a Construction Boom

A surge of financial flows is to be expected as aid arrives following a natural disaster. Physical asset replacement involving the supply of capital items (such as fishing boats and nets) that can be imported (either from overseas or from elsewhere within the country) is relatively easy and can be arranged as assistance in kind. This type of aid often does not involve financial transactions. Indeed, fishery equipment was replaced quite rapidly and, by and large, quite effectively in all three countries. However, large scale reconstruction is different. A large proportion of reconstruction usually involves the replacement of houses, buildings, bridges, roads, and other infrastructure. This effort leads to a sudden increase in demand in local markets for materials and labor inputs.¹⁶ Thus the negative economic shock due to the initial disaster is often followed by a positive “demand shock” in the construction sector. In other words, there is often a construction sector boom.

A boom of this nature usually leads to cost increases. The magnitude of the demand shock depends on the scale of the disaster and the consequent size and pace of the reconstruction program. The degree to which costs increase depends on both the size of this demand shock and on the supply responsiveness of the inputs needed in the construction sector.

If the effects of a disaster are small relative to the size of the national economy, the supply of inputs that the construction industry needs (both materials and labor) will tend to be relatively elastic. More generally, this will tend to be the case both for specific categories of construction materials in abundant supply and for relatively unskilled labor. If all the inputs in demand are internationally “tradable” and can be imported at more or less world prices, the extra demand will probably not lead to major price or cost increases. Many construction materials are indeed tradable—i.e., can be easily imported from world markets—at more or less unchanged prices. For these materials, the impact on world prices of even large increases in demand caused by a local construction program would tend to be small.¹⁷

However it is rarely the case that all of the inputs needed in a reconstruction program are in elastic supply. Because the supply of some non-tradable factors (skilled labor is a good example) is usually quite inelastic in the short run, their prices tend to rise as demand increases. Clearly, the stronger the boom in local construction, the greater will be the inflationary impact.¹⁸ Shortages of non-tradable factors are likely to be more marked in the short run than in the long run.¹⁹ In the case of skilled workers for example, this is true for two reasons: first, because over time, more skilled workers can usually be recruited to ease the shortages; and, second, because in the medium term, increasing numbers of unskilled workers can learn specialized construction skills and will thus expand the supply of skilled labor available. But many factors, particularly specialized types of labor, must often be supplied from local or at least domestic sources (and are therefore “non-traded”).²⁰

A. Construction Cost Escalation

The different experiences in Indonesia, Sri Lanka, and Thailand demonstrate how the interaction between higher demand and supply elasticities took place in different places following the tsunami. In Aceh, Indonesia, the cost of building a new 36 square meter house increased from an initial estimate of US$3,000 to around US$5,000 by end-2005. The increase was driven both by rising labor costs—somewhat moderated by the increased willingness of labor to move into the province of Aceh following the establishment of peace— and, even more importantly, by price increases in domestically sourced building materials such as timber.²¹ According to USGAO (2007: 17):

World Bank data provided by USAID show that construction costs increased as a result [of the increased demands for construction and labor]. For example, between October 2004 and October 2005, the average wage for bricklayers, plumbers, and construction supervisors in Aceh increased 55 percent, 72 percent, and 81 percent, respectively.

Overall:

…some key project activities in Indonesia and Sri Lanka— particularly its signature projects intended to generate increased visibility for U.S. assistance—have experienced increased estimated costs, are behind initial schedules, and in the case of Indonesia have been reduced in scope. (USGAO 2007: 31)

In Sri Lanka, too, total construction costs for houses planned for tsunami-affected families rose quickly. Costs were estimated to have risen by 30–50% by August 2005, and, by September 2006, had increased over initial estimates by 60–80% or more. While prices of some materials did increase in Sri Lanka, cost increases were mainly driven by higher wages for skilled labor (such as carpenters, painters, and masons) whose wages doubled in some locations (Weerakoon et al. 2007). Skilled construction labor was in scarce supply in many tsunami-affected locations so workers had to be brought in from outside the affected areas. While the tsunami destroyed the livelihoods of many people and created local unemployment, most of the unemployed people (e.g., local fishermen, farmers, small traders, etc.) had little or no construction skills.

In contrast, in Thailand construction costs actually declined during 2005 following the tsunami (Nidhiprabha 2007: 11) explains this phenomenon by emphasizing the role of local factor supply elasticities due to unemployment and excess capacity in the depressed construction sector of Thailand when the tsunami hit:

The reconstruction activities certainly increased demand in the affected regions for construction materials and labor. This was seen in the opening of a large number of construction material shops in the affected areas. However, higher demand did not lead to price increases. Here it is important to note that the tsunami-affected areas were not very far from the metropolitan Bangkok region, and the overall reconstruction activity was small relative to the size of the Thai economy. What was particularly important was that the higher demand came in the context of a depressed construction sector at the national level, reflecting the overall slowdown in economic activity, which was tending to push prices down. There was considerable excess capacity in the main input markets for construction. Substantial excess capacity in the steel industry led to declining prices of steel products used in construction, while prices of wood and wood products rose less than five per cent over prices in December 2004. Even though higher oil prices exerted some upward pressure on most materials, prices of essential raw materials such as cement and steel remained subdued during the reconstruction period. Overall, the magnitude of the demand effect was not sufficient to increase prices because there was an elastic supply of construction inputs.

Thus the particular economic circumstances in Thailand, which had yet to fully recover from the 1997 economic crisis, meant that inflation in construction costs in the tsunami-affected areas in Thailand were only moderate.

B. Dutch Disease and Reconstruction Following Disasters

This discussion of the impact of the local construction booms following the Asian tsunami has marked similarities to issues discussed in the well-known “Dutch Disease” literature. Whenever a particular sector in a particular economy experiences a marked boom, the demand for inputs used in that sector (both factors of production and materials) tends to increase. This increased demand, in turn, tends to cause negative impacts for other industries that compete for the inputs used in the booming sector. The increased prices of inputs raise costs and reduce profitability in the competing (non-booming) industries. The resulting negative impact on the non-booming sectors is known as “Dutch Disease,” sonamed after the experience in the Netherlands of de-industrialization in the wake of large inflows of export revenues from North Sea Oil in the late 1970s. The broad lesson is that in general, when countries receive large capital inflows, including foreign aid flows, expenditure is often concentrated in certain sectors. These sectors sometimes experience a marked boom while competing non-booming sectors may need to deal with negative Dutch Disease effects which flow from the resulting cost increases.²² The cost increases observed in the construction sectors following the Asian tsunami are a reflection of these Dutch Disease effects associated with absorption of financial inflows into the regional economy.

Certain features associated with this phenomenon of aid funds inflow to finance construction following natural disasters have important policy implications. Suppose, for simplicity, that two main types of inputs—tradable (imported) goods and non-tradable (domestic) goods²³—are required to support a construction boom (and, more generally, asset replacement following the loss of assets as a result of the tsunami). Suppose, also, that these inputs are used in fixed proportions. Given world prices of imported tradable goods, a given unit of foreign currency will buy a fixed quantity of imported inputs irrespective of the exchange rate of the recipient country.²⁴ But the amount of domestic non-tradable inputs that a unit of foreign currency can purchase depends on both the nominal exchange rate and on the domestic price of those inputs. The cost escalations which reflect the Dutch Disease are closely associated with the local prices of these domestic (non-traded) inputs.

If the nominal foreign exchange rate is fixed, the amount of local construction that can be financed for any given amount of foreign aid (say, $1 million) will be lower and the higher the domestic rate of inflation will be. ²⁵ The country's exchange rate policy, therefore, becomes an important matter to consider. A policy of propping up the nominal exchange rate by “leaning against the wind” in foreign exchange markets (as, for example, appears to have occurred in Sri Lanka following the tsunami) makes it much harder to fund rehabilitation or reconstruction programs with any given amount of foreign aid.²⁶ By contrast, domestic inflationary pressures can be partially mitigated by trade liberalization which tends to reduce the costs of tradable goods (including imported intermediate goods used in construction and other aid activities).

Distributional issues also need to be considered. The sensitive matter of the "fair use" of aid often arises in the wake of a disaster. It is a fact of life that some people tend to benefit more than others when a Dutch Disease type phenomenon occurs. Because of this, allegations that local traders said to be "monopolists" had indulged in activities believed to be "profiteering" and were therefore "exploiting disaster victims" are not unusual. In fact, it is true that following a disaster such as the tsunami, the local inflation of prices for inputs in short supply (such as skilled labor and certain materials) can create something of a bonanza for suppliers of these inputs. Inflation therefore tends to create a redistribution of construction-targeted aid funds, sometimes seen as an unfair windfall gain, to suppliers of these inputs. The result, within any given budget, is that plans about the scale of construction need to be revised downwards when the costs of construction rise, and, consequently, the expectations that had been raised amongst aid-beneficiary groups tend to be disappointed.²⁷

Given these problems there is surely a case for relieving supply bottlenecks by encouraging more imports, including the import of skilled labor. This approach would help reduce cost pressures and support faster construction programs. Further, an expanded reconstruction program would inject much-needed funds into the depressed local economy (see the comments in Box 1 earlier) and would help generate jobs for unemployed local people, many of whom often lack the skills to participate in the construction boom. The extra expenditure from both imported skilled labor and locally employed labor would in turn have a wider multiplier effect, lifting overall demand in the local economy. This was observed to be an important aspect of the revival of the regional economy in Aceh and Nias (Indonesia) following the tsunami.

C. Trade-Off: Pace and Amount of Reconstruction

How quickly should reconstruction proceed in the wake of a disaster? Should planners aim, as many locally-affected people often prefer, to repair the damage and to build new houses, schools, and roads as quickly as possible? Or is it better to go more slowly, and to aim to "build back better,” as many donors decided to do in Indonesia and Sri Lanka following the Asian tsunami?

Those who prefer the more measured "build back better" approach argue that if too much construction is undertaken too quickly, the regional construction boom is likely to impose unacceptable burdens on over-stretched local administrative, technical, and economic systems. They point to the risks of severe localized Dutch Disease effects as inflationary pressures mount, and also to the risks of leakages of funds if contracting processes are not carefully managed. The argument in favor of phasing in reconstruction projects over time is that if the demand for inputs is allowed to increase in a measured way, supplies of these inputs will be more elastic. The result, it is suggested, will be that both cost increases as well as the leakage of construction funds will be correspondingly lower. A more phased approach also allows time for training programs to be provided to local people in at least low-skilled construction activities which has both a cost reducing effect and a job-creating effect.²⁸ The overall result of this approach is that more capital assets can be replaced for a given amount of funding. However, a slower pace of reconstruction imposes various costs: it delays the creation of the flow of valuable services from the capital assets so that services are foregone for a longer time. Ideally, a balance needs to be struck between the high costs associated with a rapid pace of capital asset replacement and the losses due to delayed reconstruction. It is best that a program of reconstruction be planned to allow for the costs and benefits associated with different reconstruction projects. This way, rankings can be established on an economically and socially sound basis.

There are also political economy factors to bear in mind when considering the pace of reconstruction. Foreign aid donors do not hang around forever. Unless funds are put to good use quickly the funds may be diverted to other activities as donors' priorities change. There is some evidence that this happened in the use of donor funds in Indonesia and Sri Lanka following the Asian tsunami. Domestic funding can also be diverted as well. Delays in reconstruction may lead to funds being used in ways that do not really meet the needs of the worst affected groups. In other words, the costs of delays in reconstruction may fall largely on poor and politically weak groups. This issue is particularly important because a financing gap can lead to a rationing of funds. This indeed appears to be the case in Sri Lanka where reconstruction activities were heavily concentrated in politically-favored regions dominated by the majority community.

Finally, there are two points that need to be noted. One is that the negative effects of the Dutch Disease phenomenon during a construction boom following a natural disaster should not be exaggerated. The most direct negative impact of a localized construction boom would be on competing industries outside the local reconstruction areas (probably also in the construction sector in nearby regions). But this negative effect is likely to be relatively shortlived.²⁹ By its very nature, this type of boom is temporary. It reflects activities that rehabilitate the productive base of the economy including key economic infrastructure such as roads, bridges, and ports. A boom of this kind provides lasting benefits that enhance the future profitability of all sectors across the local economy. The main challenge for policymakers is to maximize the benefits that flow from the boom without dissipating reconstruction funds in ways that lead to too high windfall gains for people who were not directly affected by the disaster.

Secondly, it needs to be remembered that cost increases of the kind discussed above are inevitable. They need to be expected and budgeted for when estimating funding requirements for construction programs. Unless this is done, funding gaps will emerge. Indeed, it is surprising that the international disaster management industry apparently did not anticipate this situation in the wake of the Asian tsunami in December 2004.

Download this Paper [ PDF 152KB| 27 pages ].

Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [0] comment(s) for this entry. Post a comment.