3. Economic impact assessment

However, even where significant research regarding the economic costs of particular events has been undertaken, the poor quality of underlying data and inconsistent methods used across various events have tended to raise doubts about the findings of the studies.

Australia's Bureau of Transport Economics (BTE) found considerable variation in the methods used to estimate the economic costs of past events, and recommended a framework to facilitate future estimation of disaster costs. In general terms, BTE's suggested approach is to analyse losses arising from a hazard event in three broad categories-tangible direct, tangible indirect, and intangible (comprising the direct and indirect intangible costs).

Direct costs, which are typically the easiest to estimate, are those that arise as immediate consequences of the event itself. These costs are almost entirely associated with damage to houses, business premises, stock, etc.

Indirect costs are associated with the flow-on effects of a natural hazard. These range from losses due to business disruption to the additional cost faced by insured households who lose their no-claims bonus as a consequence of making a hazard-related claim. In an input-output modelling context, indirect costs are felt downstream as a result of either direct or other indirect costs.

To illustrate, consider a scenario from the weather bomb event. During the storm many businesses in Thames' Pollen Street sustained flood damage, both to stocks of materials and finished goods and to business property. The cost of this damage represents the direct cost of the weather bomb. In some instances, businesses were forced either to close or to reduce their level of activity, driving two related impacts. Firstly, and perhaps obviously, flood-affected businesses lost sales revenue during the time that they were operating below normal capacity. Secondly, these businesses were unable to serve their customers, many of whom were unaffected by flooding and were attempting to conducting business as usual. Plausibly, the consequence for these dependent businesses and households is that they were forced to travel further afield, hence incurring greater transportation costs than would otherwise have been the case. Both of these two secondary impacts fall within the boundaries of the indirect costs category.

It is worth noting that the sign of the net effect of a hazard event can not always be determined a priori with any certainty. In the example described above, demand for additional transportation services and petrol will have a positive impact on businesses supplying those products. Construction and cleaning services are often in demand following such an event, and this new demand is obviously of benefit to those who supply these services.

The availability of central government funding following a natural hazard event can have a large bearing on the size of the net direct and indirect impact. If, for instance, central government funds are provided to fully cover the replacement of destroyed property, the net impact to the local community may even be positive. Even in this case however, it is clear that the event is not costless; but rather that the cost is being met, if only partially, from outside the local economy.

The third category of BTE's framework is intangible costs. Intangible costs are defined as those for which there is no market, and include ongoing effects on health, household disruption and loss of memorabilia. These costs are almost certainly the hardest to quantify, and for this reason are not considered in this analysis. Injury resulting from the event, which would produce an intangible economic effect, was judged by TCDC and Civil Defence to be particularly low (none was recorded, other than one fatality).

After first providing some context for the weather bomb in terms of the Thames-Coromandel economy, the remainder of this section discusses the direct and indirect costs of the event.

The Thames-Coromandel area economy currently employs around 8,500 full-time equivalents (FTEs) (see Figure 9). The largest sectors in the Thames-Coromandel District Council (TCDC) area in terms of employment are retail and manufacturing (mostly wood and food processing), which collectively employ around 3100 FTEs. The construction, hospitality, business services and health & education sectors are also significant employers.

Over the past six years, Thames-Coromandel employment has grown more quickly than that of both the Waikato region and New Zealand as a whole. Since 1997, TCDC employment has risen 18.0 %, while Waikato and New Zealand employment climbed 12.7 % and 14.4 %, respectively.

In the context of the wider Waikato region, though, the Thames-Coromandel economy is relatively small, with employment in the TCDC area constituting less than 7 % of total Waikato employment. However, the size of the community that is directly serviced by the local economy is larger than the employment data might suggest. Of the 254,000 people living in the Waikato region, 26,300, or 10.4 %, live in the TCDC area. The gap between the employment and population data suggests strong linkages between the TCDC economy and those of surrounding districts, and in particular, that the TCDC area is a net importer of goods and services.

High levels of imports tend to lessen the impact of negative and positive shocks on the local economy. The greater the linkages with surrounding economies, the more likely an exogenous decline in demand in the local economy can be met by a decline in imports. Similarly, the more a local economy relies on external goods and services, the more likely a positive shock to the local economy will be passed on to surrounding districts.

The economic flows characterising the TCDC economy were estimated from national flows using the employment data outlined above and population census data. These TCDC economic flows are summarised in the table below.

Table 3: TCDC & NZ: selected economic flows

Note: TCDC data is estimated from secondary sources, rather than directly measured, and thus should be regarded as indicative.

Source: NZIER/GNS

The analysis that follows is based on data gathered from the two surveys and from information on insurance payouts.

As noted above, direct impacts are limited almost entirely to property damage, and this can be categorised as insured and uninsured damage. This damage can be further broken down into business and household losses.

The Insurance Council of New Zealand (ICNZ) estimate that the value of claims made as a result of the weather bomb across all of New Zealand was around $21.5 million. Of this value, around $8 million of claims are estimated to relate to the TCDC area.

Although no direct information about the split of household and business losses was available from the insurance data, these can be estimated using the survey results. According to the survey, the sum of insured losses for household respondents was $2.9 million, while for business respondents it was $0.5 million, suggesting a household/business split of 0.84/0.16. [Note the value of household loss excludes a $500,000 vehicle loss claim, which has been removed as an outlier from the derivation of the household/business split.] Applying this split to the total value of insured losses provided by ICNZ provides a total value of insured (property damage) losses to households in the TCDC area of around $6.7 million; similarly the total value of insured business losses is around $1.3 million.

The value of reported uninsured losses from respondents to the household survey was $0.7 million, while uninsured business losses were reported to have been around $0.2 million. One approach to scaling this data to represent population losses, is to use the ratio of insured losses according to the survey, to insured losses according to ICNZ. As noted above, ICNZ losses are estimated to be around $8.0 million, while insured losses from the sample survey (for both businesses and households) were reported to be around $3.5 million. A simple weighting factor relating population data to sample response can thus be estimated at around 2.3. This in turn provides an estimate of total uninsured losses to the TCDC area from the weather bomb of around $2.1 million.

In addition to these impacts are the agency response costs-i.e. the costs involved with repair to damaged drains, culverts and roads-that were incurred as a direct consequence of the weather bomb. Environment Waikato estimate that agency response costs for the TCDC area were around $3.1 million. [The Weather Bomb 21 June 2002: Final Technical Report, Environment Waikato, July 2002, p. 23.] While some of these costs will relate to replacement of damaged structures, it is assumed that they mainly relate to the labour costs associated with clearing drains and reinstating road access.

Estimates of the direct costs of the weather bomb are summarised in Table 4.

Table 4: Direct costs of the weather bomb on the TCDC area

Note: Table data may not add due to rounding

Source: NZIER/GNS

It is difficult to put these losses in context without an idea of the value of the total asset base of the TCDC area. Although no official measures of regional capital stocks exist, an estimate of the value of the asset stock of the TCDC area can be made by equating the value of TCDC's gross operating surplus presented in Table 3 above to the return on the area's asset base. Statistics New Zealand capital stock figures suggest that the ratio of gross operating surplus to productive capital stock is typically around 12 %. This rate of return implies that TCDC's asset base has a value of around $1750 million. Thus, the total value of damage caused by the weather bomb represents around 0.6 % of TCDC's pre-event asset base.

It is worth noting that this comparison effectively assumes that all losses, both insured and uninsured, are borne by the TCDC economy. The contra position is to assume that it is only the uninsured losses that are borne locally-that is, that insurance acts as an effective means of spreading local losses across New Zealand-wide premium payers. Under this scenario, the damage as assessed above represents around one-tenth of one percent of TCDC's asset base.

The true picture is likely to sit somewhere between these two extremes, and will depend on the extent to which insurance companies price policies to account for local risk. It seems plausible in the case of the TCDC area, which has been prone to a series of floods, that insurance premiums for TCDC property have been adjusted upwards to account for the additional risk inherent in covering that property. TCDC residents thus bear a greater degree of risk than does the average New Zealand policy-holder. The greater the size of this upward adjustment, the greater the extent to which insured losses are borne by the local economy.

Although this assessment is clearly crude, it is hoped that it provides an indication of the relative size of the event.

As noted above, the indirect costs associated with a natural hazard event relate to the flow-on effects of the direct impact of that event. In the case of the weather bomb, these flow-on effects take three main forms:

1. business disruption losses arising from flooding and/or property damage,
2. the potential impact on insurance premiums arising from the event, and
3. the second- and subsequent-round effects of the above and the direct effects. These costs typically affect businesses and households that are dependent in some way on businesses that were themselves directly affected by the event.

Information regarding business disruption gathered via the business survey described above was mixed. Revenue losses were reported by 39 respondents, and totalled $202,725. However, 15 respondents reported that their business increased as a result of the weather bomb, with a total increase in sales of $261,850. Thus the survey results suggest that, of those surveyed, the net impact of the weather bomb on business sales was nearly $60,000 (positive).

Although this result may seem counter-intuitive at first glance, it would be misleading to suggest that this represents the true net impact to TCDC businesses of the weather bomb. Many of the negative impacts of the event on businesses, particularly damage to property, are missing from this comparison, which instead relates solely to the impact on TCDC business turnover. Increases in turnover following a hazard event will almost certainly occur during the clean-up and reconstruction phase in specific firms, particularly those in the building, roading and general construction sectors.

A natural hazard event will have a negative impact even on those that are insured against its damage. Insurance policy-holders will almost certainly be required to pay an excess in the event of a claim, and some-those that haven't been in similar circumstances before-will lose their no-claims bonus.

These impacts relate to a single event, and their significance in terms of the local economy will depend on the proportion of affected households and businesses. IAG insurers, which covered around 56 % of all weather bomb claims under their State Insurance and NZI brands, estimate that there were 4443 weather bomb related claims throughout all of New Zealand. Around 1650 of these were made from the TCDC area, which by simple extrapolation suggests that a total of around 3000 weather bomb claims were made against all insurers by TCDC residents. [That is, 1650 (the number of IAG claims) divided by 55% (IAG's share of weather bomb claims) equals 2945.]

IAG have also indicated that the average excess payable on their claims was $150. Thus, the cost borne by insured TCDC households and businesses via excess payments totals around $450,000.

It is more difficult to gauge the extent to which TCDC residents lost any no-claims bonus as a result of making weather bomb-related claims. In part, this will depend on the degree to which TCDC residents have made prior claims-clearly for those that have, a no-claims bonus is not applicable.

The discussion of earlier claims raises the issue of the impact on insurance premiums of repeated claims resulting from similar causes. As noted above, households and businesses that are prone to flood damage will almost certainly face higher premiums as a consequence of repeated claims. At the extreme, insurers will simply refuse to continue to cover riskier properties, with the obvious consequence that the owners of those properties will bear all the inherent risk and consequent costs. Environment Waikato suggested in the follow-up to the weather bomb that, although insurance companies met most, if not all, weather bomb claims, they "may opt to cancel or refuse to renew policies for properties located within high flood risk zones (unless the risks are reduced)". [The Weather Bomb 21 June 2002: Final Technical Report, Environment Waikato, July 2002]

The impact of repeated events, particularly in terms of the costs and availability of insurance cover, can not be attributed to the weather bomb alone. However, the issue raises the point that simply looking at the impact of the weather bomb in isolation may lessen the apparent impact that flood and storm damage has on the local community over time. In short, the overall impact of successive events in the TCDC area is greater than the sum of the individual impacts of those events due to the reasons suggested above. Although it may require a degree of forward thinking, adopting a longitudinal framework involving assessment of the incremental impact of successive hazard events (as well as the individual costs of each event) is likely to provide a fairer picture of the longer-term cost of repeated events.

The final series of impacts, which have not yet been touched on, are the flow-on effects arising from any of the above direct and indirect impacts. For instance, TCDC households that face higher insurance premiums are forced to adjust their (collective) budget to accommodate this increase. This effectively means doing without some other less essential products. If these products were purchased locally, then a local provider is faced with lower sales, which in turn can have implications for his purchases, demand for labour, etc. Similarly, flooding prevented some businesses from opening in the days following the weather bomb, which in turn may have disrupted households or other business's purchasing patterns and costs.

We have used a regional version of NZIER's computable general equilibrium (CGE) model to estimate this final category of the weather bomb's indirect costs. For the purposes of this analysis, a 9-sector, 9-commodity model and social accounting matrix was used. [General notes regarding the model are attached in the appendices. More detail is available upon request from the authors.]

Perhaps unsurprisingly, these impacts were found to be negligible. Industry output, factor demand and household welfare (effectively a measure of the cost of an average household consumption bundle relative to household income) were found to be affected by less than 1 %, which in the instance of a regional New Zealand CGE model is probably close to the margin of error.

There are three main reasons why the flow-on impacts of the weather bomb are likely to have been minimal:

1. The duration of the event was relatively short. Although some businesses were still returning to normal weeks after the event, many had returned to 'business as usual' within a few days. Although electricity outages were widespread in the area, in most cases service was resumed within a day. [WEL Energy reported that the longest outage time was 14.5 hours.] As a result, the impact of the weather bomb in terms of disruption to day-to-day economic flows was relatively mild.
2. The severity of the weather bomb in terms of its direct economic costs was also relatively mild. As noted above, the weather bomb is estimated to have affected just one-tenth of one percent of the total value of TCDC's asset base. The impact on households' insurance premiums was of a similar size relative to the size of total TCDC household consumption. [That is, the payment of no-claims bonuses by TCDC residents was around $450,000, while total TCDC household is around $400 million (see Table 3).]
3. TCDC is in effect a small, borderless economy. The consequence of this is that any disruptions to local supply that did occur (i.e., where a business was forced to close for an extended period as a result of the weather bomb) that could not be met via alternative local suppliers, could be met relatively easily from suppliers in the non-affected surrounding districts.

The table below presents a summary of the value of the total economic impact of the weather bomb on the TCDC economy.

Table 5: Total costs of the weather bomb on the TCDC area

Source: NZIER/GNS

Losses resulting from the weather bomb in the wider South Waikato area were relatively insignificant compared to those in the TCDC area and compared to the size of the South Waikato area and economy. Household survey respondents reported around $400,000 of insured losses and around $50,000 of uninsured losses (compared to TCDC survey responses of $2.9 million and $0.7 million, respectively). Unfortunately, the insurance data collected did not provide the basis on which to derive a value for population losses for the South Waikato area. However, from the magnitude of the survey responses, it seems reasonable to assume that both direct and indirect losses are relatively insignificant in relation to the South Waikato economy.