In Singapore, the Pulau Bukom Manufacturing Site will pivot from a crude oil-based product slate towards new, cleaner fuels. Photo credit : Straits Times

It’s time to move on, the debate on carbon tax is over

It’s great that Singapore Workers Party MP Jamus Lim has kicked off the debate on carbon tax and dividend. For those of you who get the impression that it’s not clear cut because it seems like there are two valid arguments on both sides, I would like to take this opportunity to deconstruct one recent counter-argument opposing the carbon tax, either wholly or for incremental price increase.

This argument’s main claim is that a carbon fee and dividend is a net loss for low income families because the losses from the lost opportunity of economic activity outweigh the gains from the cash payment.

The main structure of the argument claims that carbon taxes induce a “distortionary” effect on a specific sector of the economy, this distortion reduces incentives for investment in that sector, since investments create jobs and increase productivity this reduces job creation, wage growth and the higher energy prices increase operating costs for all businesses.

The thing that seems to be in agreement on both sides is support for an increase in public investments in decarbonization. So to fully critique this argument would be to compare a carbon tax + public investments with public investments funded from broad based taxes. Either of these policies could be effective at incentivizing the decarbonization switches what is more decisive is the total size S$bil of the investments from public and private sector. This post is not ambivalent however and attempts to make an ambitious claim that a carbon tax would either be a poor trade-off of environmental and economic objectives of the state or at worse an ineffective decarbonization policy instrument. Neither of these claims is well supported by evidence nor do the represent the consensus of economists on the topic.

The case for raising the carbon tax is compelling as a market based solution. From a conventional economics perspective there are several non controversial considerations that make it difficult to support the claims from the post.

So let’s go one by one

The post doesn’t cite any references to support claims

There is a ambiguous reference to “*Tax economists* widely agree that narrow-based taxes like carbon taxes raise taxes far less efficiently than broad-based labour/consumption taxes, particularly due to the way it hurts the economy”

— which tax economists? The consensus among economists on carbon taxes as an effective policy tool for reducing emissions = mitigating climate change is fairly well established. One good read written for layman audience from a reputable institution is the IMF Fiscal Monitor Oct 2019[1]. That report references one of the most widely cited economic analysis of emissions reduction policy the Stern Review 2008[2]. Sources that I am aware of that make similar blanket statements favoring broad based taxes over carbon (Pigovian) taxes are think tank institutes that are known to have funding from oil and gas affiliates such as the Heritage Foundation which makes almost this exact same claim about carbon taxes in their 2020 Economic Freedom report[3]. Another place to find arguments against carbon tax and for broad based tax is the advocacy groups for “fair tax” which was a US based 501c3 founded by a Houston (oil and gas hub) construction businessman Leo Linbeck Jr [4].

Circumstantial links suggesting ulterior motives however is not sufficient evidence to refute an argument. So let’s get into the substance

All fiscal activity “distorts” the economy, sometimes in a good way

Yes, carbon taxes, or any government fiscal action be it taxes or spending distorts the economy from what it would have been otherwise. That is a design feature not a bug. Taxes on income distort the labor market by adding a friction between businesses and employees and taxes on capital distort capital markets by putting an additional premium on capital ownership. Government spending on healthcare and education distort the market by making the effective cost lower and more accessible than it would be in a lassie-fair system. So blanket use of the word “distortion” is void of meaning and just a stand in for the effect of government fiscal policy on the economy and to use it in a negative is only a restatement of a preference for small government size as measured as tax burden (= total tax collected / GDP).

You have to tax something and better to tax externalities than things that are good for the economy (labor, consumption)

So a more meaningful question is which distortions are also net beneficial for the economy? Generally spending is good because it puts money into the economy, and taxes not so good with the exception of Pigovian taxes. I am assuming a balanced budget as I believe the author also implicitly assumes. Pigovian taxes are designed to discourage an activity which is known to have a harmful public hazard “externality” that the market doesn’t effectively price on its own. The environmental is a classic example of externalities but there are numerous others such as gambling, smoking, alcohol consumption, or bubble tea. Pigovian taxes internalize these externalities and create market incentive to invest in reducing the economic losses to the real economy (natural, human capital) from these activities. For the case of carbon emissions, a carbon tax penalizes emitting carbon dioxide into the atmosphere. I will not explain here why that is something that needs to be done hopefully by now that is general public knowledge and the author also agrees that this is an objective for the state.

While the author eludes to two potentially competing objectives he doesn’t state explicitly how to resolve the trade off the long term cost to the real economy in terms of natural capital from carbon emissions and the short term cost to the economy from the Pigovian tax. Fortunately economists have been studying this for a while and the term “social cost of carbon” is the collective cost on society of unmitigated climate change. Because climate change occurs so far into the future the social cost of carbon is almost completely determined by the extent to which one discounts future economic activity vs the present typical values range from $40-$80/ton US[2] which is close the the recommended range for the price level for the carbon tax. So the carbon tax price is designed to exactly this — to resolve the time inconsistency between future costs of climate change and current lost opportunity costs from utilization of carbon intensive energy sources.

Yes taxes may reduce some investments but dividends create new ones

The situation though is more benign because the economy is approximately closed system. For every dollar collected in carbon tax there is another dollar spent in the form of a carbon dividend. While it’s true the carbon tax is likely to discourage new investments in carbon intensive projects, the carbon dividend is a source of funding for new investments in non-carbon intensive sectors. Either the individual spends on new economic activity that they would not have otherwise which attracts new entrepreneurs and private capital or they save it and it gets re-deployed by fund managers. There is a complication if an economy has large asymmetrical FDI or balance of payments but for Singapore the symmetry is in their favor for now and it would be quite a stretch to argue that a carbon fee and dividend would have any effect on this asymmetry. The new investments though while they may sustain the economy in a steady state, they may not have any net improvement. Real long term monthly wage growth is determined only by total factor productivity growth, fixed capital formation, monthly hours per worker, and labor/capital substitution. Of each of these only total factor productivity has potential to increase without bounds, all of the others have some theoretical upper limits. Although the fixed capital formation limit could be debated the author also highlighted productivity as the ultimate measure of long term improvement for the economy. The author is correct to say that in the long run government spending on R&D and public investments in infrastructure and public services- healthcare, education have better long term productivity growth prospects, especially for an advanced economy like Singapore, than direct payments.

Singapore’s petrochemicals industry is <7% of GDP and <1% of the workforce

So the critical question then is whether there is something special about Singapore that the lost opportunity from investments in carbon intensive projects is >> the economic benefit from new investments from the revenue collected — either from direct payments or government spending. The author correctly identified which industry would be affected — oil and gas and energy production, but then he starts to make some slippery slope claims of cascading ripple effects on the rest of the economy without citing any evidence to support these links. While the petrochemical industry in Singapore looks big when you view it up close like I have working for the said polluters in their facility. I can testify the physical scale of their fixed capital is impressive to see up close. That’s not how we measure economic impact however. Another area that is noteworthy is that they are one of the highest labor productivity. For those familiar with economics this is an obvious consequence of the large fixed capital and the labor capital substitution effect. Aside from these two dimensions, the industry is only <7% of GDP and <1% of the workforce. So it’s tough to argue this slippery slope for such a small fraction of the economy. Even if the whole industry caved, which it wouldn’t from just a carbon tax, we are talking about 25,000 workers. That could easily be absorbed from any number of modestly sized new investments in other industries (healthcare, education, digital services, sustainability..). The reason is precisely because of the large differences in labor productivity. Employment potential per $ of investment is inversely related to labor productivity. Higher productivity industries require larger investments compared to lower productivity industries to create the same number of new jobs. That’s why TOTAL factor productivity is a better yardstick than labor productivity.

Investments in CCUS also create jobs

How would the industry respond to the emissions price increase?

It’s true that the prospects are not great but this is more a function of global developments outside of Singapore. They indicated from their reports that they would shift investments into decarbonization (CCUS) projects. While this does have an implication for the ultimate long term cost of energy, in the short run those investments are ultimately job creation using the same skills as the industry as a whole, so very likely the net job loss would be << 25,000 and might be more in the short run. Upstream would definitely lose out this is without a doubt but it’s less gloomy for petrochemicals.

S$150/year is enough to cover any incremental electricity price increase for low income households

Could a sharp downturn in the petrochemical industry send ripple effects to other industries via transmission of energy costs? This sounds scary until you look at the numbers. The industries with substantial energy component of operation costs is the petrochemicals, >20% by my rough estimates[5]. Even for manufacturing energy costs as % of total open is not so large, and keep in mind a fully priced carbon tax may only increase energy costs by +15–20%, so we are talking about <2% increase in opex for a company with 10% of energy as opex which is pretty high figure. Most companies are less than this. So it’s a bit of a stretch to claim there will be ripple effects on total economy wide on <2% of opex. Also, the claims of lost jobs in petrochemicals and the claim of ripple on the whole economy is double-dipping. Either the cost is passed on or it is absorbed as a loss in the sector. It can be a diluted mix of the two effects but not the worst case for both. What would residents expect to pay? Worst case +20% price increase for a typical HDB $80/month electric bill amounts to <$200/year which is within striking distance of the $150/year dividend so in a diluted mixed effect scenario (more likely) with price impact <20% it’s either a wash or households are in-the-money.

The IMF, Stern, Jamus Lim and Louis Ng have all done their homework on this one and the carbon tax and dividend has wide public support. There is hardly a more clear consensus on a policy improvement than that for raising the carbon tax in Singapore. Let’s save our energy for real debates.




Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Taylor Hickem

Taylor Hickem


Applied research, engineering, and projects for solutions to sustainable cities. SG Green New Deal