Skip to main content

Advertisement

SpringerLink
Log in

Econometric Evidence on Forest Ecosystem Services: Deforestation and Flooding in Malaysia

  • Published:
Environmental and Resource Economics Aims and scope Submit manuscript

Abstract

Governments around the world are increasingly invoking hydrological services, such as flood mitigation and water purification, as a justification for forest conservation programs in upstream areas. Yet, rigorous empirical evidence that these programs are actually delivering the intended services remains scant. We investigate the effect of deforestation on flood-mitigation services in Peninsular Malaysia during 1984–2000, a period when detailed data on both flood events and land-use change are available for 31 river basins. Floods are the most common natural disaster in tropical regions, but the ability of tropical forests to mitigate large-scale floods associated with heavy rainfall events remains disputed. We find that the conversion of inland tropical forests to oil palm and rubber plantations significantly increased the number of days flooded during the wettest months of the year. Our results demonstrate the importance of using disaggregated land-use data, controlling for potentially confounding factors, and applying appropriate estimators in econometric studies on forest ecosystem services.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Notes

  1. Ogden et al. (2013, p. 8444) note that the original table in Kiersch (2000) did not include any supporting references.

  2. A pioneering cross-country study on the economics of natural disasters by Kahn (2005) included an analysis of flood occurrence, but it focused on issues other than the effects of land use. It obtained flood data from the Emergency Events Database (EM-DAT; http://www.emdat.be), which is widely used but has shortcomings that include separate flood events being recorded as a single one and smaller flood events being underreported in developing countries (Jonkman 2005, p. 153).

  3. The other two are the states of Sabah and Sarawak on Borneo.

  4. These surveys do not cover Sabah and Sarawak.

  5. If our objective had been to investigate underlying hydrological processes, then we would have needed to include potentially confounding factors directly as covariates instead of using fixed effects to sweep away their effects.

  6. One type of overfitting in panel models occurs when fixed effects for cross-sectional units are included at a level that is more finely disaggregated than necessary for identifying the effect of interest; it causes standard errors to be underestimated but does not affect the consistency of parameter estimates (Ritschl 2009). In our models, this type of overfitting would occur if RBMUs could be classified into homogeneous groups, in which case fixed effects should be included for groups instead of individual RBMUs. The RBMUs do not fall into obvious groups, however, as they differ greatly in area, topography, and other fixed characteristics. Moreover, this type of overfitting occurs only if the number of fixed effects is large relative to the number of observations, which was not the case in our models (31 RBMU fixed effects, 636–716 observations).

  7. Including a large number of controls can also cause multicollinearity, but given our objective this was a lesser evil than omitted-variables bias. Multicollinearity inflates standard errors, but it does not bias parameter estimates (Kennedy 2002, pp. 193–194). Hence, it contributes to a conservative estimation strategy by reducing the risk of overestimating parameter significance.

  8. Similarly, given that the Department of Agriculture conducts the land-use surveys using the same procedures in all parts of Peninsular Malaysia, there is no apparent reason to expect measurement error in the 1984–1985, 1997–1998, and 2004–2005 surveys to be nonrandom across RBMUs or land uses.

  9. Clustering is a second reason that we did not use the AIC or BIC to guide model specification: these statistics should not be used when data have a clustered structure (Hilbe 2011, p. 69).

  10. The 10-month gap from 1 year to the next meant that we did not need to cluster the standard errors at an even lower frequency, e.g., the entire sample period for a given RBMU, because the gap interrupted the serial correlation process between years.

  11. Descriptive statistics differed little for the 716-observation sample used in the models for number of floods.

  12. The parameter estimates cannot be directly compared, as they are interpretable as semi-elasticities in the Poisson model and marginal effects on log odds ratios in the logit model.

  13. These effects refer to months when flooding occurred, not all months in the sample, as they are based on coefficients from the zero-truncated component of the hurdle model.

  14. One RBMU contained a much higher percentage of wetland forest than the others: 59 % as of 1984, with the next highest percentage being 32 %. Results changed little if we excluded that RBMU from the sample. For example, the coefficients on oil palm, rubber, and wetland forest in model (3) when estimated for that sample were 0.280 (\(P =\) 0.001), 0.275 (\(P =\) 0.002), and 0.312 (\(P =\) 0.003).

  15. We also considered the effect of using samples that defined wet months in ways other than the core wet season. In one variant, we expanded the sample by extending the wet season to include the preceding and following months. One would expect this to reduce the absolute value of the regression coefficients on the land-use variables, given that the coefficients reflect mean monthly effects on days flooded and that extending the sample causes those effects to be averaged over a period that includes months when there is less rainfall, less saturated soils, and thus less flooding. (In the extreme case of no flooding at all, the effects of land use on days flooded during those months would necessarily be zero.) This is what we found; for example, the coefficients on oil palm and rubber were reduced to 0.147 and 0.155, respectively (\(P < 0.05\) for both). In a second variant, we ignored seasonality and included months with rainfall in the 70th percentile. This yielded a sample with 670 observations, similar to the samples in Table 3. Coefficients on the land-use variables retained their signs but became more significant and smaller: the positive ones became less positive, and the negative one became more negative. The coefficients on oil palm and rubber were 0.192 and 0.212, respectively (\(P < 0.001\) for both).

  16. Flooding could be affected by time-varying factors other than the three included in model (3). One example is regional income; another is forest management, which under the Malaysian constitution is the responsibility of state governments. To investigate the influence of such factors, we added annual trends for the nine states in the sample to model (3). With the exception of the “other” land-use variable, which remained negative and insignificant, coefficients on all the other land-use variables increased. For example, the coefficients on oil palm and rubber were 0.385 and 0.602, respectively (\(\hbox {P} < 0.01\) for both).

  17. \(2 \times (0.268 \times 1.11) / (0.01 \times 3343) = 0.018\).

References

  • Alila Y et al (2009) Forests and floods: A new paradigm sheds light on age-old controversies. Water Resour Res 45: doi:10.1029/2008WR007207

  • Angrist JD, Pischke JS (2009) Mostly harmless econometrics: an empiricist’s companion. Princeton University Press, Princeton

    Google Scholar 

  • Bennett G et al (2013) Charting new waters. Forest Trends, Washington

    Google Scholar 

  • Bradshaw CJA et al (2007) Global evidence that deforestation amplifies flood risk and severity in the developing world. Glob Change Biol 13:2379–2395

    Article  Google Scholar 

  • Brauman KA et al (2007) The nature and value of ecosystem services. Annu Rev Environ Resour 32:67–98

    Article  Google Scholar 

  • Bruijnzeel LA (2004) Hydrological functions of tropical forests. Agric Ecosyst Environ 104:185–228

    Article  Google Scholar 

  • Cameron CA, Trivedi PK (2013) Regression analysis of count data, 2nd edn. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • Carlson KM et al (2012) Committed carbon emissions deforestation and community land conversion from oil palm plantation expansion in West Kalimantan Indonesia. Proc Natl Acad Sci USA 109:7559–7564

    Article  Google Scholar 

  • Department of Statistics (1981) General report of the population and housing census 1980. Kuala Lumpur, Malaysia

    Google Scholar 

  • Department of Statistics (1992) General report of the population and housing census 1991. Kuala Lumpur, Malaysia

    Google Scholar 

  • Department of Statistics (2001) General report of the population and housing census 2000. Kuala Lumpur, Malaysia

    Google Scholar 

  • Drainage and Irrigation Department (2010) National hydrological networks. http://h2o.water.gov.my/v2/fail/Invstations/Inventory.html

  • Drainage and Irrigation Department (2013) Dam information. http://www.water.gov.my/our-services-mainmenu-252/dams/dam-informations-mainmenu-834?lang=en

  • Energy Commission Malaysia (2013) Electricity supply industry in Malaysia: performance and statistical information 2011. Putra Jaya, Malaysia

    Google Scholar 

  • FAO (2010) Global forest resources assessment 2010: main report. Food and Agriculture Organization of the United Nations, Rome

    Google Scholar 

  • FAO and CIFOR (2005) Forests and floods. Food and Agriculture Organization of the United Nations, Rome and Center for International Forestry Research Bogor, Indonesia

  • Ferraro PJ et al (2012) Forest figures. Rev Environ Econ Policy 6:20–44

    Article  Google Scholar 

  • Ferreira S, Ghimire R (2012) Forest cover socioeconomics and reported flood frequency in developing countries. Water Resour Res 48: doi:10.1029/2011WR011701

  • Ferreira S et al (2013) Does development reduce fatalities from natural disasters? New evidence for floods. Environ Dev Econ 18:649–679

    Article  Google Scholar 

  • Gutiérrez-Vélez RH, DeFries R (2013) Annual multi-resolution detection of land cover conversion to oil palm in the Peruvian Amazon. Remote Sens Environ 129:154–167

    Article  Google Scholar 

  • Hilbe JM (2011) Negative binomial regression, 2nd edn. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • Huber PJ (1967) The behavior of maximum likelihood estimates under nonstandard conditions. In: Neyman LM (ed) Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, vol 1. University of California Press, Berkeley, pp 221–233

    Google Scholar 

  • Jonkman SN (2005) Global perspectives on loss of human life caused by floods. Nat Hazards 34:151–175

    Article  Google Scholar 

  • Kahn ME (2005) The death toll from natural disasters: the role of income, geography, and institutions. Rev Econ Stat 87:271–284

    Article  Google Scholar 

  • Kennedy P (2008) A guide to econometrics, 3rd edn. Blackwell, Malden, Massachusetts

    Google Scholar 

  • Kiersch B (2000) Land use impacts on water resources: a literature review. In: Calder IR (ed) Land-water linkages in rural watersheds. Food and Agriculture Organization of the United Nations, Rome

    Google Scholar 

  • Kinzig AP et al (2011) Paying for ecosystem services–promise and peril. Science 334:603–604

    Article  Google Scholar 

  • Koh LP, Ghazoul J (2010) Spatially explicit scenario analysis for reconciling agricultural expansion forest protection and carbon conservation in Indonesia. Proc Natl Acad Sci USA 107:11140–11144

    Article  Google Scholar 

  • Koh LP et al (2011) Remotely sensed evidence of tropical peatland conversion to oil palm. Proc Natl Acad Sci USA 108:5127–5132

    Article  Google Scholar 

  • Kramer RA et al (1997) Ecological and economic analysis of watershed protection in Eastern Madagascar. J Environ Manage 49:277–295

  • KTA Tenaga (2003) Updating of condition of flooding in Malaysia–main report. Drainage and Irrigation Department, Kuala Lumpur

  • Lambert D (1992) Zero-inflated Poisson regression with an application to defects in manufacturing. Technometrics 34:1–14

    Article  Google Scholar 

  • Li H et al (2007) Demand for rubber is causing the loss of high diversity rain forest in SW China. Biodivers Conserv 16:1731–1745

    Article  Google Scholar 

  • Millennium Ecosystem Assessment (2005) Ecosystems and human well-being: synthesis. Island Press, Washington

  • Moulton BR (1986) Random group effects and the precision of regression estimates. J Econom 32:385–397

    Article  Google Scholar 

  • Mullahy J (1986) Specification and testing of some modified count data models. J Econom 33:341–365

    Article  Google Scholar 

  • Ogden FL et al (2013) Effect of land cover and use on dry season river runoff runoff efficiency and peak storm runoff in the seasonal tropics of Central Panama. Water Resour Res 49: doi:10.1002/2013WR013956

  • Pattanayak SK, Kramer RA (2001) Worth of watersheds. Environ Dev Econ 6:123–146

  • Pattanayak SK, Wendland KJ (2007) Nature’s care: diarrhea, watershed protection and biodiversity conservation in Flores, Indonesia. Biodivers Conserv 16:2801–2819

  • Pattanayak SK et al (2010) Show me the money. Rev Environ Econ Policy 4:254–274

    Article  Google Scholar 

  • Ritschl A (2009) Spurious significance of treatment effects in overfitted fixed effect models. London School of Economics

  • Van Dijk AJM et al (2009) Forest-flood relation still tenuous. Glob Change Biol 15:110–115

    Article  Google Scholar 

  • Vincent JR, Hadi Y (1993) Malaysia. In: National Research Council (ed) Sustainable agriculture and the environment in the humid tropics. National Academy Press, Washington

  • Vincent JR, Mohamed Ali R (2005) Managing natural wealth: environment and development in Malaysia. Resources for the Future, Washington

    Google Scholar 

  • White HL (1980) A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica 48:817–838

    Article  Google Scholar 

  • Wong IFT (1971) The present land use of West Malaysia 1966. Ministry of Agriculture and Lands, Kuala Lumpur

    Google Scholar 

  • Wooldridge JM (2002) Econometric analysis of cross section and panel data. MIT Press, Cambridge

    Google Scholar 

  • Zeger SL, Liang KY (1986) Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42:121–130

    Article  Google Scholar 

  • Ziegler AD, Fox JM, Xu J (2009) The rubber juggernaut. Science 324:1024–1025

    Article  Google Scholar 

  • Zulkifli Y et al (2004) Runoff estimate from the North Selangor Peat Swamp Forest. In: Abdul Rahim N (ed) Water: forestry and land use perspectives, Technical Documents in Hydrology No. 70, International Hydrological Programme. UNESCO, Paris

    Google Scholar 

Download references

Acknowledgments

This study was funded by the Global Environment Facility through the United Nations Development Programme (MAL/04/G31). Additional support was provided by the Government of Malaysia, through the Ministry of Natural Resources and Environment and the Forest Research Institute Malaysia, and by the Center for International Forestry Research. The cooperation of the Departments of Agriculture, Drainage and Irrigation, and Statistics of the Government of Malaysia is gratefully acknowledged as are helpful comments by M Jeuland, D Richter, and E Sills.

Author information

Authors and Affiliations

  1. Duke Global Health Institute, Nicholas School of the Environment and Sanford School of Public Policy, Duke University, Durham, NC, 27708, USA

    Jie-Sheng Tan-Soo

  2. Nicholas School of the Environment and Sanford School of Public Policy, Duke University, Durham, NC, 27708, USA

    Subhrendu K. Pattanayak & Jeffrey R. Vincent

  3. Forest Research Institute Malaysia, 52109, Kepong, Selangor, Malaysia

    Norliyana Adnan & Ismariah Ahmad

Authors
  1. Jie-Sheng Tan-Soo
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Norliyana Adnan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ismariah Ahmad
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Subhrendu K. Pattanayak
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Jeffrey R. Vincent
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jeffrey R. Vincent.

Additional information

This paper has not been submitted elsewhere in identical or similar form, nor will it be during the first 3 months after its submission to the Publisher.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Tan-Soo, JS., Adnan, N., Ahmad, I. et al. Econometric Evidence on Forest Ecosystem Services: Deforestation and Flooding in Malaysia. Environ Resource Econ 63, 25–44 (2016). https://doi.org/10.1007/s10640-014-9834-4

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10640-014-9834-4

Keywords

Search

Navigation