Skip to main content
Log in

An Economic Analysis of Earth Orbit Pollution

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

Abstract

Space debris, an externality generated by expended launch vehicles and damaged satellites, reduces the expected value of space activities by increasing the probability of damaging existing satellites or other space vehicles. Unlike terrestrial pollution, debris created in the production process interacts with firms’ final products, and is, moreover, self-propagating: collisions between debris or extant satellites creates additional debris. We construct a formal model to explore private incentives to launch satellites and to mitigate space debris. The model predicts that, relative to the social optimum, firms launch too many satellites and choose technologies which create more debris than is socially optimal. We discuss remediation strategies and policies, and demonstrate that Pigovian taxes can be used to internalize the debris externality.

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

Access this article

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

Instant access to the full article PDF.

Fig. 1

Similar content being viewed by others

Notes

  1. A very general economic taxonomy of types of goods is straightforward: private goods are rival (in use) and excludable; club goods are non-rival and excludable; public goods are non-rival and non-excludable; and common goods are rival and non-excludable.

  2. Hardin observed over-use of common resources by individuals; in the case of orbital space, overuse is driven by nations.

  3. Unlike many terrestrial common resources, technological progressivity generally out-weighs geography in accessing earth-orbit. A nation’s geographic proximity to space is roughly equivalent to any other nation’s; technological capability is the essential separating feature, which can only be overcome by developing the required technological knowledge and skills, or by contracting with an entity that has a presence in orbit. Contrast this with access to other common resources such as the earth’s oceans: even some technologically-advanced landlocked countries do not have ocean-faring navies or direct access to fisheries. Thus, the rules of access are, in this instance, driven by technology.

  4. In fact, space was, and is, not a pristine common in the generally accepted sense of pristine. Enormous quantities of naturally occuring debris pass through orbital space in the form of meteriods and space dust.

  5. Awareness of the need for an institutional framework for the use of space trace back to the first half of the 20th century. A detailed review of this earlier literature is given in Lyall and Larsen (2009).

  6. http://www.spacecolorado.org/news/2011-state-of-the-satellite-industry-report-shows-growth.html.

  7. http://www.futron.com/upload/wysiwyg/Resources/Reports/SSIR2010.

  8. http://www.ucsusa.org/nuclearweaponsandglobalsecurity/spaceweapons/technicalissues/ucs-satellite-database.html.

  9. http://earthobservatory.nasa.gov/Features/OrbitsCatalog/.

  10. http://www.ucsusa.org/nuclearweaponsandglobalsecurity/spaceweapons/technicalissues/ucs-satellite-database.html.

  11. http://www.orbitaldebris.jsc.nasa.gov/library/EducationPackage.

  12. http://www.orbitaldebris.jsc.nasa.gov/newsletter/pdfs/ODQNv13i2.

  13. http://www.orbitaldebris.jsc.nasa.gov/library/EducationPackage.

  14. http://www.oosa.unvienna.org/pdf/limited/AC105C12011CRP14E.

  15. Liou et al. (2010).

  16. Wright (2007), and also McKnight (2010) who observes that the destructiveness of the impact depends on both the mass of the objects as well as the encounter geometry.

  17. History of On-Orbit Satellite Fragmentations, 14th Edition, Orbital Debris Program Office, NASA/TM2008.

  18. Limiting Future Collision Risk to Spacecraft: An Assessment of NASAs Meteoroid and Orbital Debris Programs; http://orbitaldebris.jsc.nasa.gov/newsletter/pdfs/ODQNv14i2.

  19. Kessler (1991). This is analogous to the “shallow lake” problem explored in Brock and Starrett (2003) or a bit more popularly, a “black swan” event.

  20. Limiting Future Collision Risk to Spacecraft: An Assessment of NASAs Meteoroid and Orbital Debris Programs, Committee for the Assessment of NASAs Orbital Debris Programs, Aeronautics and Space Engineering Board, Division on Engineering and Physical Sciences, September 1, 2011.

  21. It is important to note that location is not to be taken literally in our model. Distance is stylized and represents the loss of satellite service quality which is experienced as a consequence of buying service from a provider which is less-than ideal (i. e. not “located” immediately in the neighborhood of a consumer.)

  22. It is important to note, simply for the purpose of accuracy, that substantial demand for satellite services exist for military and para-military organizations in the United States and elsewhere. At least some of the demand of these institutions is self-supplied i.e., these organizations have their own satellites in orbit.

  23. See Tirole (1988) for a discussion on the appropriateness of a circular model when analysing the number of firms in a differentiated product industry.

  24. We abstract throughout from naturally occurring orbital debris.

  25. Note that in period one, the firm has two fixed costs: \({r>0}\) (the launch cost), and \(F>0\), (the maintenance cost). However, in period two the firm has only one fixed cost, \(F>0\).

  26. See Salop (1979) or Tirole (1988) for further discussion on the derivation of the demand curve in a circular model.

  27. Our model implies the risk of a collisional cascade, since an increase in debris creation (or satellite loss risk) induces more launches which further increases the probability of more collisions. Thus, there is a positive feedback loop which, without mitigation, can lead to a “Kessler syndrome” (see Kessler and Cour-Palais 1978).

  28. The formula for the social planner is the same as in (9) with the modification that the social planner can endogenously choose parameter \(\phi \).

  29. Kessler and Cour-Palais (1978).

  30. Liou (2011).

  31. See also Macauley (2003) for a policy discussion on remediating debris.

  32. McKnight (2010), among other things, provides an overview of active debris programs and proposals.

  33. Note that \(r = h(\phi _{soc})\) when the social planner can select \(\phi \).

References

  • Brock WA, Starrett D (2003) Managing systems with non-convex positive feedback. Environ Resour Econ 26:575–602

    Article  Google Scholar 

  • Demsetz H (1967) Toward a theory of property rights. Am Econ Rev 57(May):347–359

    Google Scholar 

  • Fehr E, Kremhelmer S, Schmidt KM (August 2008) Fairness and the optimal allocation of ownership rights. Econ J 118

  • Hardin G (1968) The tragedy of the commons. Science 162(3859):1243–1248

    Article  Google Scholar 

  • Kessler DJ (1991) Collisional cascading: The limits of population growth in low earth orbit. Adv Space Res 11(12):347–351

    Google Scholar 

  • Kessler DJ, Cour-Palais BG (1978) Collision frequency of artificial satellites: the creation of a debris belt. J Geophys Res 83(A6):2637–2646

    Google Scholar 

  • Liou JC (2011) A parametric study on using active debris removal for LEO environment remediation. Adv Space Res 47(11):1865–1876

    Google Scholar 

  • Liou JC, Johnson NL (2006) Risks in space from orbiting debris. Science 311(5759):340–341

    Google Scholar 

  • Liou JC, Johnson NL, Hill NM (2010) Controlling the growth of future LEO debris populations with active Debris removal. Acta Astronaut. 66(5–6):648–653

    Article  Google Scholar 

  • Lyall F, Larsen PB (2009) Space law: a treatise. Ashgate, UK

    Google Scholar 

  • Macauley M (2003) Regulation on the final frontier. Regulation 26(2):36–41

    Google Scholar 

  • McKnight D (2010) Pay Me Now or Pay Me More Later: Start the Development of Active Orbital Debris Removal Now. In: Paper presented at the advanced Maui optical and space surveillance technologies conference

  • Ostrom E (1990) Governing the commons: the evolution of institutions for collective action. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • Ostrom E, Burger J, Field CB, Norgaard RB, Policansky D (1999) Revisiting the commons: local lessons, global challenges. Science 284(5412):278–282

    Google Scholar 

  • Salop SC (1979) Monopolistic competition with outside goods. Bell J Econ 10(1):141–156

    Article  Google Scholar 

  • Schwartz E (2011) “Waste MGMT”, The Best American Science and Nature Writing, Houghton, Mifflin, Harcourt

  • Tirole J (1988) The theory of industrial organization. The MIT Press, Cambridge

    Google Scholar 

  • Wright D (2007) Space debris. Phys. Today 60:35–40

    Article  Google Scholar 

Download references

Acknowledgments

We are indebted to David Sappington, Donald Kessler, Bill Gibson, Michael Ash, Nicholas Johnson, Scott Pace, Brian Weedon, Henry Hertzfeld, Heidi Garrett-Peltier, James Boyce, Ceren Soylu, Kevin Crocker, Jerry Duvall, and the seminar participants in the Environmental Working Group, PERI Institute, University of Massachusetts, Amherst, for their useful comments on early drafts.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Nodir Adilov.

Additional information

Peter J. Alexander and Brendan M. Cunningham: The views presented here reflect the views of the authors, and do not reflect the views of the Federal Communications Commission or the United States Naval Academy. No government resources were used in producing this draft, and all data are from publicly available sources.

Appendices

Appendix 1: Proof of Proposition 1

$$\begin{aligned}&\displaystyle \frac{\partial L_{Com}}{\partial t} = \frac{1}{2t} \sqrt{\frac{(\beta + (1 - k \phi ))}{(1 - k \phi )(r + F + \beta (1-k \phi ) F)}} > 0 \end{aligned}$$
(17)
$$\begin{aligned}&\displaystyle \frac{\partial L_{Com}}{\partial \beta } = \frac{1}{2\sqrt{L_{Com}}} \cdot \frac{t}{(1 - k \phi )} \cdot \frac{r + F k \phi (2 - k \phi )}{(r + F + \beta (1-k \phi ) F)^2} > 0\end{aligned}$$
(18)
$$\begin{aligned}&\displaystyle \frac{\partial L_{Com}}{\partial \phi } = \frac{1}{2\sqrt{L_{Com}}} \cdot \frac{k \beta t (r + F + 2 \beta F (1 - k \phi ) + F (1- k \phi )^2) }{[(1 - k \phi )(r + F + \beta (1-k \phi ) F)]^2} > 0\end{aligned}$$
(19)
$$\begin{aligned}&\displaystyle \frac{\partial L_{Com}}{\partial k} = \frac{\partial L_{Com}}{\partial \phi } > 0 \end{aligned}$$
(20)
$$\begin{aligned}&\displaystyle \frac{\partial L_{Com}}{\partial r} = \frac{1}{2\sqrt{L_{Com}}} \cdot \frac{- (1 - k \phi )(\beta + (1 - k \phi ))t}{[(1 - k \phi )(r + F + \beta (1-k \phi ) F)]^2} < 0\end{aligned}$$
(21)
$$\begin{aligned}&\displaystyle \frac{\partial L_{Com}}{\partial F} = \frac{1}{2\sqrt{L_{Com}}} \cdot \frac{- (1 - k \phi )(1 + \beta (1 - k \phi ))(\beta + (1-k \phi ))t}{[(1 - k \phi )(r + F + \beta (1-k \phi ) F)]^2} < 0 \end{aligned}$$
(22)

Appendix 2: Proof of Corollary 1

\(\frac{\partial L_{soc}}{\partial t} = \frac{1}{2} \frac{\partial L_{Com}}{\partial t}> 0,\,\frac{\partial L_{soc}}{\partial \beta } = \frac{1}{2} \frac{\partial L_{Com}}{\partial \beta } \!>\!0,\,\frac{\partial L_{soc}}{\partial \phi } = \frac{1}{2} \frac{\partial L_{Com}}{\partial \phi } > 0,\,\frac{\partial L_{soc}}{\partial k} \!=\! \frac{1}{2} \frac{\partial L_{Com}}{\partial k} > 0,\frac{\partial L_{soc}}{\partial r} \!=\! \frac{1}{2} \frac{\partial L_{Com}}{\partial r} < 0\), and \(\frac{\partial L_{soc}}{\partial F} = \frac{1}{2} \frac{\partial L_{Com}}{\partial F} < 0\).

Appendix 3: Proof of Proposition 2

Comparing \(L_{Com}\) and \(L_{soc}\) yields \(L_{soc} = \frac{1}{2} L_{Com} < L_{Com}\).

Appendix 4: Proof of Proposition 3

Differentiating \(\pi _i\) defined in (13), with respect to \(\phi _i\) yields:

$$\begin{aligned} \frac{\partial \pi _i}{\partial \phi _i}&= - h'(\phi _i) - \frac{\beta k}{L} \left( \frac{t}{((1- k \bar{\phi })L)^2} - F\right) + \frac{2 k \beta t}{L^3 (1 - k \bar{\phi })^2} \end{aligned}$$
(23)
$$\begin{aligned}&= - h'(\phi _i) + \frac{ k \beta t}{(1 - k \bar{\phi })^2 L^3} + \frac{\beta k F}{L} > 0 \end{aligned}$$
(24)

This implies that firm \(i's\) profit is strictly increasing in \(\phi \). Therefore, it is optimal for each firm \(i\) to choose \(\phi _i = \phi _H\) regardless of what other firms choose. This completes the proof.

Appendix 5: Proof of Proposition 4

First, we note that the problem defined in (15) has a solution because the value function \(V\) is continuous in \(\phi \) on its closed and bounded domain. Differentiating \(V\) with respect to \(\phi \), while noting that \(\frac{\partial V}{\partial L} = 0\), yields:

$$\begin{aligned} \frac{d V}{d \phi } = L(\phi ) h'(\phi ) + \frac{\beta k}{4 L(\phi ) (1 - k \phi )} (t - 4 F (1 - k \phi ) (L(\phi ))^2) \end{aligned}$$
(25)

To prove that \(\phi _{soc} < \phi _H\), it suffices to show that \(\frac{d V}{d \phi } |_{\phi _H} > 0\). Evaluating (25) at \(\phi _H\) yields \(\frac{d V}{d \phi } |_{\phi _H} = \frac{\beta k}{4 L(\phi _H) (1 - k \phi _H)} (t - 4 F (1 - k \phi _H) (L(\phi _H))^2)\). This expression is positive if \(t - 4 F (1 - k \phi _H) (L(\phi _H))^2 > 0\).

$$\begin{aligned}&t - 4 F (1 - k \phi _H) (L(\phi _H))^2 \\&\quad =t - 4 F (1 - k \phi _H) \frac{(\beta + (1 - k \phi _H)) t}{4 (1- k \phi _H) (h(\phi _H) + F + \beta (1 - k \phi _H) F)} \\&\quad =\frac{(h(\phi _H) + (1 - \beta ) k \phi _H F) t}{h(\phi _H) + F + \beta (1 - k \phi _H) F} > 0 \end{aligned}$$

Next, we compare the launch rates. According to Corollary 1, \(L_{soc} (\phi _{soc}) < L_{soc} (\phi _{H})\) because \(\phi _{soc} < \phi _H\). Furthermore, according to Proposition 2, \(L_{soc} (\phi _{H}) < L_{Com} (\phi _{H})\). Therefore, \(L_{soc} (\phi _{soc}) < L_{Com} (\phi _{H})\). This completes the proof.

Appendix 6: Proof of Proposition 5

Suppose the regulator selects a variable tax rate \(\gamma ^*(\phi ) = \{ \gamma (\phi ) = 0 \) if \(\phi = \phi _{soc},\,\gamma (\phi ) = s\) if \(\phi \ne \phi _{soc} \}\). Because \(s\) is the maximum consumer surplus value for the market (as introduced in Sect. 2.1.1), \(\phi = \phi _{soc}\) is the only choice for firms that could yield non-negative profits. Thus, any firm that chooses to launch a satellite would select \(\phi = \phi _{soc}\). Next, suppose the regulator sets a per launch tax rate \(w^* = 3(h(\phi _{soc}) + F + \beta (1 - k \phi _{soc}) F )\). Then, the tax schedule equals \(T^* = 3(h(\phi _{soc}) + F + \beta (1 - k \phi _{soc}) F ) + \gamma ^*(\phi )\). Equation (5) from Sect. 2.2 describes the number of launches in the competitive market for a given \(\phi \). Then, under the tax schedule \(T^*\), the number of launches in the competitive market would equal:

$$\begin{aligned} L_{Com}&= \sqrt{\frac{(\beta + (1 - k \phi _{soc}))t}{(1 - k \phi _{soc})(h(\phi _{soc}) + T^* + F + \beta (1-k \phi _{soc}) F)}} \\&= \sqrt{\frac{(\beta + (1 - k \phi _{soc}))t}{(1 - k \phi _{soc}) (4) (h(\phi _{soc}) + F + \beta (1-k \phi _{soc}) F)}}\\&= \frac{1}{2} \sqrt{\frac{(\beta + (1 - k \phi _{soc}))t}{(1 - k \phi _{soc})(h(\phi _{soc}) + F + \beta (1-k \phi _{soc}) F)}} = L_{soc} \end{aligned}$$

Thus, \(L_{Com} = L_{soc}\) and \(\phi _{com} = \phi _{soc}\) under the tax schedule \(T^*\).

Rights and permissions

Reprints and permissions

About this article

Cite this article

Adilov, N., Alexander, P.J. & Cunningham, B.M. An Economic Analysis of Earth Orbit Pollution. Environ Resource Econ 60, 81–98 (2015). https://doi.org/10.1007/s10640-013-9758-4

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10640-013-9758-4

Keywords

Navigation