You are on page 1of 18

A Strategic Explanation for Unilateral GHG Regulation:

Peter Davis (10593)


2

Introduction:

As the international climate conference in Copenhagen approaches, the difficulty of


negotiating and implementing a comprehensive international agreement for climate
change mitigation has become apparent. There has been considerable progress over
the last decade towards the formation of an international climate protection regime.
However, for those who see climate change as a cataclysmic and potentially
irreversible crisis, progress at an international level has been frustratingly slow. This
frustration is often perceived to be the driving force behind the proliferation of
numerous national and sub-national climate change mitigation policies.

The causes and impacts of sub-global voluntary action to reduce the emissions of
greenhouse gases (GHGs) is increasingly a subject of interest to social scientists.
Whether these programs will ultimately have a significant and durable impact on
climatic change depends on the underlying rationale that is driving their development.
If these programs are irrational anomalies, as they are sometimes depicted by the
economic literature, then they might be predicted to peter out as their costs rise. If,
however, these programs are locally rational—not requiring costly philanthropic
behaviour—then these programs might be dynamically stable or even dominant.

The first contribution of this paper to the existing literature is to formally model a
category of individual behaviour that directly or indirectly reduces the individual’s
GHG emissions. The behaviour of interest in this paper is the decision to divest from
long-term capital stocks that might be exposed to regulatory risks under a global
climate change regime. The divestment behaviour explained by this model is
reinforced by ancillary benefits to the investor from emissions reductions, and by a
signalling effect, which can help investors resolve the principal agent problem.

The second contribution of this paper extends the results from the private sector to
consider the implications for local and regional public policy decisions. This section
introduces the possibility of local ancillary benefits from GHG emission reduction.
Alone, these benefits from emission reduction could justify limited local climate
protection programs, especially if other regions have enacted similar programs. But,
3

if the private sector is known to deal poorly with systematic risk and long time
horizons, then the local public sector has an even greater incentive to intervene.

The conclusion of this analysis is not that decentralized GHG regulation is a substitute
for a comprehensive climate change regime. Because of carbon leakages, these
unilateral initiatives will have a limited direct impact on short term global GHG
emissions. Nonetheless, decentralized GHG regulation is a significant and growing
phenomenon. As the proliferation and intensification of decentralized GHG
regulation gains momentum these programs can be expected to have an increasing
impact on country level payoffs from an internationally negotiated regulatory regime.
Furthermore, in a finite world, the carbon leakages from a marginal regulator can be
expected to decrease as the proportion of regulators increases. For these reasons,
decentralized GHG regulation is nontrivial; rather it is the driving force of global
climate change mitigation.

Literature Review:

The majority of political and economic scholarship on climate change mitigation


deals with the issue of negotiating, implementing and enforcing a comprehensive
international regime. Contributions to this debate generally focus on the problem of
designing an efficient, yet feasible, mechanism for achieving a socially optimal level
of GHG emissions: one which distributes costs and benefits ethically and efficiently
across space and time. The emphasis on the international level has overshadowed a
marginal scholarship on the local decision making processes that can directly, or
indirectly, influence GHG emissions and global climate change negotiations.

Even among economists who consider unilateral regulation, it is commonly argued


that local action to reduce GHG emissions will have negligible, or even perverse,
effects on aggregate global emissions. This argument is formalized by Hoel (1991).
The game theoretic model Hoel uses is two representative individuals who make
decisions regarding a purely public good, given the expected responses of the other
individual. The conclusions of this game theoretic approach to public goods were
strongly pessimistic for unilateral emissions regulation.
4

Not only would the degree of unilateral regulation be below the globally optimal
level, but the effects of unilateral emissions reduction on global emissions would be
diluted by international carbon leakage. Hoel also argued that unilateral emissions
reductions would reduce the bargaining power of the regulating country in
international negotiations, thereby diluting the future global regulatory regime.

Hoel’s bargaining power argument is perhaps an oversimplification of the process of


international negotiations and that allowing for costly punishment can strengthen the
unilateral regulator’s bargaining position. On the subject of carbon leakages,
however, a number of empirical studies demonstrate the significance of this
phenomenon. In a simulation, Felder and Rutherford (1993) found that marginal
leakages to be in the order of 25 percent. Babiker (2005) empirically measured
carbon leakages and found them to as large as 50 to 130 percent. Weiner (2006)
attributes such large leakages to the unexpectedly fast shift of emissions towards Asia.
This shift is visually presented below in Figure 1, while Figure 2 shows the
insignificant impact of European regulation on global GHG emissions.
5

Figure 1: Global CO2 Emissions Centre of Gravity (1970-2005)

Figure 2: The Impact of European Emission Reductions on Global Emissions

Nonetheless, the last decade has seen a proliferation of unilateral actions to reduce
GHG emissions. Since its inception in 1993, the Cities for Climate Protection (CCP)
campaign has grown to involve more than 650 local governments, representing 15
percent of global GHG emission.1 Lutsey and Sperling (2008) estimated that 53
percent of the U.S. national population lived in a state or city that had a target for
GHG emission reduction in 2007. In China, plans are underway for a new city of half
a million people in Dongtan that will generate zero contribution to global warming
(Keohn, 2008). Further evidence of the scale of unilateral actions is available from a
number of sources including Rabe (2004), Betsill and Bulkeley (2004), and Collier
and Löfstedt (1997).

Given that unilateral GHG emission reduction will not have a significant, or could
even have a perverse, effect on the level of global GHG emissions; how can the

1
www.iclei.org
6

proliferation of unilateral actions be explained? There are two categories of solutions


to this puzzle.

The first category of solutions specify that for unilateral climate change action to be
rationalized it must have an expected impact on global GHG emissions. These
solutions take two forms possible forms: individuals could have systemically incorrect
expectations; or individual behaviour could have indirect impact on global GHG
emissions through influencing the behaviour of others. These solutions require either
that unilateral actors be purely irrational agents, or quasi-rational normative agents.
Schreurs and Tiberghien (2007) present such a solution to the puzzle of consistent
European Union leadership in climate change mitigation.

The second category of solutions specify that for unilateral climate change action to
be rationalized it must have an expected utility benefit to the actor, regardless of the
impact on global GHG emissions. Do individual actors have an incentive to reduce
GHG emissions, independent of the climatic impacts of these actions? And do cities
and regions have an incentive to regulate GHG emissions in the absence of an
international climate change regime?

A solution to the puzzle of unilateral GHG regulation that falls within the category of
individually rational behaviour is much more desirable than a solution that falls in the
category of quasi-rational behaviour. Quasi-rational behaviour fails to resolve the
free-rider problem, because the appearance of emissions reductions can be as effective
as actual reductions. This creates the perverse incentive for corporations and
governments to “green-wash” their practices, rather than making real costly
transformations.

A purely rational explanation for unilateral emissions reduction is one that, in


evolutionary terms, improves the fitness of the unilateral actor. In evolutionary
models, a behavioural strategy which improves the fitness of an individual can be
expected to be dynamically stable, or even dominant. If this is the case for unilateral
climate change action, then this behaviour can be expected to replicate itself over
time. The argument of Schreurs and Tiberghien, while eloquently articulated and
heuristically interesting, fails to satisfy this condition.
7

There are few existing rational actor models that explain unilateral climate change
mitigation. Urpeleinen (2009) skirts the issue in his two-level game that shows how
experimentation by lower levels of government, like the State of California, reduces
the uncertainty of political costs and benefits at the national level. The national
leadership might therefore prefer sub-national policies. This is an effect that the
International Energy Agency has termed the Real Options Approach (IEA, 2007).
Urpeleinen fails, however, to address the root causes of what he calls the
“Schwarzenegger phenomenon”. Urpeleinen’s only rigorous support for unilateral
action is to recognize that GHG emission reduction is an impure public good—
meaning that some benefits of abatement accrue directly to the region.

Several other authors have also commented on the ancillary benefits from reducing
GHG emissions. Koehn (2008) emphasizes the importance of explicitly bundling co-
benefits of climate change action in promoting local climate change initiatives.
Koehn explores several examples of how reducing GHG emissions could potentially
have ancillary benefits at a local and regional level. These benefits include reduced
environmental degradation, diminished resource consumption and depletion, and
health-related effects. These all constitute rational explanations for why a local or
regional government would choose to regulate GHG emissions unilaterally. But, not
taking into account global aggregate emissions means that the level of regulation due
to ancillary benefits will always be sub-optimal.

The contribution of this paper is to introduce the concept of corporate sustainability—


or the risks associated with corporate insustainability—as a rational solution to the
puzzle of unilateral GHG regulation. Richardson (2009) charts the emergence and
evolution of the concepts of corporate social responsibility, socially responsible
investment and corporate sustainability. Richardson is pessimistic, however, about
the potential of corporate sustainability to have a significant impact on global
warming. This is because he perceives the movement to have lost its
ethical/ideological/normative foundation:

“For most investors, global warming is (at most) mainly a matter of financial risk or investment
opportunity. With decision-making in financial institutions dominated by fund managers and other
investment professionals focused on the ‘bottom line’, any notion that such institutions are a forum
8

of enlightened ethical deliberation among climate-conscious investors would be naive.


(Richardson, 2009:602)

Echoing Richardson, the Dow Jones Sustainability Index (DJSI) maintains that “a
growing number of investors perceive sustainability as a catalyst for enlightened and
disciplined management, and, thus, a crucial success factor”. Klaus Wellershoff,
global head of UBS Wealth Management Research, adds:

“Whether or not you agree with the view that human activity is influencing the climate system is
largely irrelevant to the investment thesis. What is important is that numerous policies to combat
the threat of global warming are converging to influence people’s behaviour, alter the risk profile
of various businesses, and improve the investment outlook for others” (Baue, 2007)

This paper demonstrates that the conception of “sustainable” as it relates to the


corporate bottom-line can have greater consequences for international climate change
mitigation than conceptions that concentrate on corporate philanthropy or altruism.
The rational expectations of firms regarding future regulation can create an incentive
for investors to unilaterally divest from carbon-intensive industries before
international negotiations are concluded. The rest of this paper formalizes the micro-
economic principals of this argument and explores the impact of divestment on the
incentives for local and global GHG emissions regulation. When local externalities
and imperfections in capital markets are considered, a strong case emerges for
unilateral GHG regulation, irrespective of the impact of such regulation on global
GHG emissions.

A Microeconomic Model of Unilateral Climate Change Mitigation:

This section presents a simple micro-economic model of individual behaviour


resulting in GHG emissions in a large society of N individuals. Babiker (2007)
provides a similar, but more complex, model of GHG emissions. In this model, a
representative firm produces output Y using two technological processes:

The inputs in both processes are capital (K) and carbon (C). The first process is
carbon intensive, while the second process is capital intensive. Carbon is a purely
variable input that must be purchased in each period and is totally consumed in the
9

production process. Capital purchases spill-over into the next period. Ownership of
capital stock represents a flexible store of wealth for an investor in the firm because
they can be divested to increase consumption at a rate:

The total wealth (W) of the investor at time t is therefore:

Investors tailor their investment portfolio to reflect their preference for carbon-
intensive and low-carbon capital investment. The investor’s utility gained from
investment is a function of the net present value (NPV) of future wealth, but also a
function of the expected variance or risk associated with different investment
decisions. The investor’s utility function can therefore be described by:

The Ω parameter of the utility function represents the individual’s risk preferences.
To make rational decisions about long-term investment the investor must have a set of
expectations about the future prices of capital in the two sectors. These expectations
can be described by three vectors:

There are several factors that determine these prices and are therefore taken into the
consideration of a rational investor. The price of carbon is determined by
environmental scarcity, by demand, and by government regulation. A rise in the price
of carbon could be triggered, for instance, by the implementation of an international
climate change mitigation regime; or by a sudden decrease in the availability of
hydrocarbons.

As a baseline, it is useful to consider the relative price of the two capital goods in a
society that does not consider future price expectations. Prices are therefore
10

determined by the equilibrium condition that the marginal revenue products of K1,
K2, and C are all equal:

Or

Given

In a static environment, therefore, an investor will be indifferent between different


portfolio compositions. There are, however, a number of external shocks that can
influence the relative prices of capital-goods in the future. Firstly technological
advancement in the production of low-carbon capital can cause the relative price of
low-carbon capital to fall. This creates a bleeding edge effect of early adaptors
paying unreasonably high costs. The bleeding edge effect might be offset by
subsidies, regulatory incentives, or by intellectual property rights that enable early
adaptors to earn rents from future technological advances.

Another potential shock to the relative price of capital-goods would be a rise in the
price of carbon. The resulting technical substitution to low-carbon technology would
cause demand for carbon-intensive technology to evaporate. The higher the ratio of
carbon-intensive capital in the individual’s investment portfolio, the more exposed the
individual is to the impacts of a change in carbon prices. A risk adverse investor
may choose to divest from carbon-intensive capital, if he expects the price of carbon
to rise in the future. This is analogous to a forward looking individual divesting from
a pyramid scheme or market bubble before its inevitable collapse.

As the international community approaches a comprehensive climate change


agreement and uncertainty in low-carbon technology decrease, the number of
investors divesting from carbon-intense capital should increase and the demand for
carbon-intense capital should begin to decline. As the price of carbon-intensive
capital decreases, risk preferring or myopic investors will increase their portfolio
11

share of carbon-intense capital and earn temporary economic rents on the cheap
capital.

Figure 3: Price and MRP of Carbon Intense Capital

MRP (Cheep Carbon)

MRP (Costly Carbon)

Probability of GHG Regulation

This process may not lead to drastic reductions in aggregate GHG emissions in the
short term, but it has the beneficial effect of discouraging new investment in carbon-
intense capital, because the supply of new capital is determined purely by its
production costs. The other effect of this process is to concentrate carbon-intensive
industries in the portfolios of risk-preferring individuals. The combination of this
factor with the existence of ancillary benefits from reducing GHG emissions creates a
strong incentive for regulation at the local and national levels. This effect is
demonstrated by a game theoretic model in the following section.

Local and Regional Regulation:


This section illustrates a dynamic model of local and regional GHG emissions
regulation. It finds that regions have an incentive to regulate GHG emissions before
the implementation of a comprehensive climate change mitigation regime.
Furthermore—contrary to the public goods model—when other regions increase their
regulation of GHGs the appropriate response is not to decrease regulation. Because of
12

the nature of the ancillary benefits, there is a strong first mover advantage in tougher
regulation.

Societal intervention in the market is efficient when the interests of individual actors
diverge from the interests of society as a whole. The function of societal intervention
is to force actors internalize the social costs of their behaviour. Intervention is called
for in the case of GHG emissions for a number of reasons.

Firstly, because the expected costs of anthropogenic climate change exceed the costs
of mitigation. This is the global public goods argument. But, this potential benefit of
regulation is diffused at the global level, so local community leaders will not
internalize the full benefits of averting climate change into their decision making
process.

There are, however, a number of social benefits from regulation that can be
internalized at the local level and regional level. These externalities are not limited to
the ancillary benefits mentioned previously. Drawing from the previous section, it is
feasible that there will be costs associated with a sudden loss of wealth that are not
internalized by the owners of capital. If a large percent of a community’s wealth and
productive capacity evaporated overnight, jobs would be lost and the community
could potentially fall into a recession. For this reason carbon-intensive investments
can be seen as having societal risks that are not internalized by the private investor.

The divergence of private and public interests is exacerbated when investors expect to
be compensated for their investments in the event of regulation. There is a well
established norm of compensation to avoid regulation that is unfairly punitive and to
smooth the transition into the new regulatory regime, avoiding some of the
aforementioned costs. This norm has been respected to some extant by most land-
reform programs, by the emancipation of serfs and slaves, and by various alcohol
prohibition programs. The pragmatic explanation for this norm is that, without
compensation, the capital interests harmed by regulatory reform have an incentive to
use their political clout to obstruct the reforms. If this is the case then private
investors may not even fully internalize the private risks of their investments.
13

A final point is that even if private capital was to internalize the cost of future price
shocks, there would still be a case for societal intervention. This is because societal
risk and time preferences may differ from those of the private sector. This argument
may be controversial with classical economists. Baumol (1968), for example, argued
that public and private discount rates should be equal. Recent scholarship, however,
appears to accept the idea that society should be more patient and risk adverse than
the private sector (Caplin and Leahy, 2004; Farhi and Werning, 2007).

Societies therefore have an incentive beyond the typical ancillary benefits to regulate
GHG emissions. The societies that choose not to regulate do so either because
increased risk and pollution is seen as an acceptable trade-off for increased activity,
because of a lack of information regarding the true costs and risks associated with
carbon-intensive capital, because of inefficient political markets, or because the
administrative costs of regulation exceed the technical capacity of governance.

Together, these barriers create a frictional cost that tends to discourage smaller and
less politically developed societies from adopting GHG regulations. The existence of
such frictional costs is confirmed by Robinson and Gore’s (2005) work on barriers to
municipal regulation in Canada; and by Collier and Löfstedt’s (1997) work on the
experience of municipal governments attempting to regulate GHG emissions in
Sweden and in the United Kingdom.

But, how will the utility of regulation change when a neighbouring society
implements stricter regulations? The classical approach suggests that the probability
of an international regime should be inversely related to the proportion of societies
with GHG emissions regulations. Hoel’s bargaining power argument suggests that
when a society regulates unilaterally it gives up emission reductions that could have
been used in negotiations to obtain concessions from other parties. I argue contrarily,
that an increase in the number of regulating societies will not only increase the
likelihood of an international regime, but will also strengthen the incentive of non-
regulating societies to introduce their own regulation.

When a society regulates unilaterally it accepts a loss of productivity relative to non-


regulating societies. This creates an economic transfer between the regulating and
14

non-regulating societies, until an international agreement is reached that forces all


societies to regulate. Regulating societies therefore have an incentive to hasten an
international agreement by punishing non-regulating societies.

Assume x percent of societies unilaterally regulate GHG emissions, and x-1 percent
of societies remain unregulated. The total economic costs of carbon leakages are a
strictly increasing function of x, because marginal carbon leakages are assumed to be
strictly positive. A coalition of regulators would therefore choose a punishment level
λ that would be strictly increasing in x. The costs of this punishment will be divided
between regulators and non-regulators:

This conceptualization creates a critical point for x (in the preceding functional form
it is found at x=0.5). Before this point, punishment is irrational because it will incur a
greater cost to the marginal regulator than to existing non-regulators; in effect
deterring regulation. After the critical-mass of regulators is reached the punishment
of a large number of regulators is concentrated on a small number of non-regulators.
From this point onwards, the incentive for non-regulators to join the coalition of
regulators is strictly increasing in x.

The process may be amplified by increasing ancillary costs of non-regulation as


carbon-intense capital moves into the remaining non-regulating regions. As x
becomes large, the few non-regulators will be flooded with dirty, risky, carbon-
intense capital. Once the ball is rolling, this system should eventually produce the
stable Nash equilibrium where all societies regulate GHG emissions (with or without
an international regime) and the societies that adapt last pay the highest price.

Concluding Remarks:

This paper proposed a possible rational explanation for local and regional
governments to unilaterally regulate their GHG emissions. Both investors and the
regions in which they invest have an incentive to minimize their exposure to
regulatory risk. Ancillary costs to society from the emission of GHGs encourage
15

governments to regulate GHG emissions beyond the private sector equilibrium. The
incentive to regulate unilaterally is even stronger if the private sector does not
adequately internalize systemic risks or if the private sector’s risk and time
preferences differ from those of the society.

--The trajectory should follow an S-curve.


1st Technology is Uncertain.. Carbon Leakages are High. Low Carbon Market is
small.. Few societies unilaterally regulate.. some societies develop real position
options.
2nd Critical Mass of Low Market is formed. Pollution problems emerge in non-
regulating regions. Low carbon technologies stabilize.

The implication of this result is that as the negotiation of an international climate


change regime, the more countries, municipalities and individual investors will divest
from carbon-intensive capital stocks. In perfectly competitive markets with good
information this behaviour should follow a continuous s-curve of divestment over
time. This behaviour will yield a comparative advantage for these regions and
individuals in the future low-carbon economy. The trade-off is that risk preferring
investors will earn rents from the risky carbon-intensive capital until an agreement is
reached.

Non-regulating municipalities and regions will benefit from increased economic


activity, but these benefits will eventually be eclipsed by the external costs and the
risk associated with carbon-intense capital. Still, the governments of these regions
might be unable to regulate due to inefficient political markets, lack of technical
capacity, insufficient awareness of the risks associated with GHG emissions, or other
barriers to regulation. This understanding of carbon leakages demands the question:
where will the carbon intense economy leak to when China regulates GHG emissions?
And then where?

Clearly, there is no long term advantage from engaging in an environmental race to


the bottom. Eventually, the world must make the transition to low-carbon technology.
This is the fundamental flaw in models that conceptualize emission abatement as a
pure global public good. The scholarship addressing regime theory and institutional
16

design is interesting and important, but it should not disregard the potential
significance of complex individual and regional behaviours in the absence of an
international regime. More theoretical work is needed to understand the determinants
of unilateral abatement, and also to understand the impacts of unilateral abatement on
international negotiations.
17

Bibliography:
Babiker, Mustafa (2005) Climate Change Policy, Market Structure, and Carbon
Leakage. Journal of International Economics, 65(2):421-45.

Baue, Bill. Citigroup, Lehman Brothers, and UBS Report on Climate Risks and
Opportunities for Investing. Sustainability Investment News, February 27th, 2007.
Available at: <http://www.socialfunds.com/news/article.cgi/2237.html>

Baumol, William (1968). On the Social Rate of Discount. The American Economic
Review, 58(4):788-802.

Betsill, Michele and Bulkeley, Harriet (2004). Transnational Networks and Global
Environmental Governance: The Cities for Climate Protection Program. International
Studies Quarterly, 48(2):471-93.

Brown, Marilyn (2001). Market Failures and Barriers as a Basis for Clean Energy
Policies. Energy Policy 29(14):1197-207.

Caplin, Andrew and Leahy, John (2004). The Social Discount Rate. Journal of
Political Economy, 112(6)

Collier, Ute and Löfstedt, Ragnar (1997). Think Globally, Act Locally? Local Climate
Change and Energy Policies in Sweden and the U.K. Global Environmental Change,
7(1):25-40.

Farhi, Emmanuel and Werning, Iván (2007). Inequality and Social Discounting.
Journal of Political Economy, 115(3):365-402.

Felder, S. and Rutherford, T.F. (1993). Unilateral CO2 Reductions and Carbon
Leakage: The Consequences of International Trade in Oil and Basic Materials.
Journal of Environmental Economics and Management, 25(2):162-76.

Hoel, M. (1991). Global Environmental Problems: The Effects of Unilateral Actions


Taken by One Country. Journal of Environmental Economics and Management 20,
55–70.

International Energy Agency (2007). Modeling investment risks and uncertainties


with real options approach—a working paper for an IEA Book: Climate Policy
Uncertainty and Investment Risk

Keohn, Peter (2008). Underneath Kyoto: Emerging Subnational Government


Initiatives and Incipient Issue-Bundling Opportunities in China and the United States.
Global Environmental Politics, 8(1):53-77.

Lutsey, Nicholas, and Daniel Sperling. (2008). America’s Bottom-Up Climate


Change Mitigation Policy. Energy Policy 36 (2): 673–685.
18

Schreurs, Miranda A., and Yves Tiberghien. (2007). Multi-Level Reinforcement:


Explaining European Union Leadership in Climate Change Mitigation. Global
Environmental Politics 7 (4): 19–46.

Rabe, Barry G., Mikael Roman, and Arthur N. Dobelis. (2005). State Competition as
a Source Driving Climate Change Mitigation. New York University Environmental
Law Review 14 (1): 1–53.

Richardson, Benjamin (2009) Climate Finance and its Governance: Moving to a Low
Carbon Economy Through Socially Responsible Financing?. International and
Comparative Law Quarterly, 58:597-624.

Robinson, Pamela, and Gore, Christopher (2005). Barriers to Canadian Municipal


Response to Climate Change. Canadian Journal of Urban Research, 14(1):102-120.
Schreurs, Miranda and Tiberghien, Yves (2007). Multi-Level Reinforcement:
Explaining European Union Leadership in Climate Change Mitigation. Global
Environmental Politics, 7(4):19-46.

Urpelainen, Johannes (2009). Explaining the Schwarzenegger Phenomenon: Local


Fruntrunners in Climate Policy. Global Environmental Politics, 9(3):82-105

Wiener, Jonathan (1998) On the Political Economy of Global Environmental


Regulation. Georgetown Law Review, 87(1):749-794.

Wiener, Jonathan (2006). Think Globally, Act Globally: The Limits of Local Climate
Policies. Duke Law School Faculty Scholarship Series. Paper 93

Velasquez-Manoff, Moises (2009) Companies Desert the Climate Deniosphere. The


Christian Science Monitor, August 28th. Available at:
<http://features.csmonitor.com/environment/2009/09/28/companies-desert-the-
climate-deniosphere/>

You might also like