Thursday, July 14, 2011

Very Low Cost Electricity -- Impacts on Commodity Prices

What impact would very low cost electricity have on commodity markets? First let's consider on commodity prices that are used for electricity generation -- mainly coal and natural gas. This is to say, if, the great scientist John Von Neumann was correct in the previous post, nuclear energy had become extremely abundant to the extent that natural gas and coal were mainly used for chemical feedstocks, then the expenditure that was used for natural gas and coal for electricity would be diverted to (costless) nuclear fuel -- or another costless source of electrical power.

However note, that the previous post calculated the savings from electricity, which should include the expenditures for coal and natural gas. This is to say, as almost all of the coal and natural gas in the United States is purchased by power utilities as inputs, and then sold to consumers through electricity as kilowatt hours, savings in electricity costs would include the savings in costs spent on natural gas and coal (interestingly, this calcualtion makes sense, the entire natural gas market in the United States in 2010 was 24.1 billion cubic feet, which at a price of $5 per cubic foot is approximately $120Bn. The price of all electricity sold in the US in 2009 was $383Bn (about 4.5 Billion megawatt/hours at 9.7 cents per megawatt -- the total expenditure on electricity includes the expenditure on natural gas and coal).

Further we will assume that electric power has become so abundant that natural gas is not even used for heating, but for chemical feedstock purposes only. This usage represents actual cost savings however (not included in electricity), so in 2010, the US according to the EIA supplied 8.2 Billion cubic feet to residential and office and commercial buildings, which at a price of $5 per cubic foot would be cost savings of $41Bn.

Natural gas in particular is used as a feedstock for chemical synthesis, as is oil -- approximately 20% of the oil consumed by the United States is used as a feedstock for petrochemicals. According to the EIA, the US supplied Industry with 6.6B cubic feet. Let's assume without the recent rise in natural gas prices over the last decade, with very low cost energy, natural gas would average long term prices from 1980-1999, of approximately $1.50 per cubic foot (much natural gas is produced along with oil, and producers of oil could supply natural gas as a side business to chemical producers). This would mean savings of ($5-$1.50)*6.6Bn = $23.1Bn

According to the EIA, only approximately 2% of coal was used for non-electricity generation purposes in 2010, so the vast majority of cost savings that are not already calculated by savings in electricity costs.

How would the price of Oil be affected by very low cost electricity?

Oil prices are determined by supply and demand. Demand: Oil on the demand side is mainly used for transportation (80% in the US) and petrochemicals (20%) in the US. As oil is a world market, the world demand for oil determines world prices, and world demand is a higher percentage for transportation (as the United States is the world's largest producer of petrochemicals).

It is not clear if a large number of cars would be converted to electrical (battery-operated if electricity was very low cost, and whether this would encourage more public transport dependent on electricity (trains, trams, etc). Currently the firm "A Better Place" is undertaking projects to convert a large percentage of cars in several countries (Israel, Denmark, Australia) to battery powered -- in light of high oil prices. With extremely low electricity costs, more battery powered vehicles is more feasible.

On the supply side, low cost electricity would assist exploration and production through lower cost drilling as well as refining and marketing, through lower electricity costs.

How much would oil prices be lowered in a world of very low electricity costs, stemming from these factors? It is very difficult to tell. Supply would be incentivitzed to increase by a small amount, while demand would likely decline, by what amount is very difficult to tell. One could say that China would be more likely to build a transportation infrastructure highly dependent on electricity, in light of its relatively low domestic reserves of petroleum, which would likely have reduced demand by a great deal throughout the decade of the 2000's. China has been cited as one of the main drivers of higher oil prices, by several commentators.

For the purposes of this analysis, oil prices with very low electricity costs are assumed to decline to levels averaging in the 1990's, of approximately $20 per barrel (this is a large assumption! But also shows how much oil prices have risen over the last decade). With prices of $20 per barrel, the US would save ($90 per barrel approximately currently - $20 per barrel)*19.5M barrels consumed per day*365 = approximately $498Bn per year.

Would the prices of non-energy related commodities such as food (grains) and metals decline with very low cost energy?

As the prices of all commodities are determined by supply and demand, one would go through the sources of supply and demand for each commodity to determine what direction the prices would move. It would be very difficult to argue that prices would actually go up with very low cost and abundant electricity. For the purposes of this analysis, I will leave out the cost savings as an impact on GDP (as this post is getting very lengthy:).

Environmental Impacts from Low Cost, Clean Energy:

An additional benefit to GDP would be from environmental impacts. Carbon production would be significantly lower, as most carbon in the atmosphere is due to coal production and oil production (coal consumption would be much lower, oil consumption would be debatable, how much it would decline, as argued above).

Oddly, GDP impacts from lower carbon would not be dramatic as of 2011, as very few companies engage in carbon trading/credits, however if some publications and scientists are to be believed -- such as the late Steven Schneider of Stanford University, a lower amount of carbon in the atmosphere is an extremely -- extraordinarily -- serious matter.

Some countries' GDP such as Norway includes environmental degradation, but the US' GDP to a large extent does not, so this would not be directly added to GDP. I will look for environmental estimates of the impact on GDP from other published sources.

Additional Technological Breakthroughs from Very Low Cost Electricity:

Very low cost electricity would likely lead to more scientific breakthroughs (besides the knowledge of low cost nuclear energy in itself) with more funds being available for scientific research grants, and lower costs to research labs (which depend on electricity as a cost and an input to experiments). This is likely a critical outcome of lower cost electricity, but I won't go into detail on this as this time as this post is getting quite lengthy.

Conclusion: Calculation of Additional Savings from Lower Oil and Natural Gas Prices:

In addition, to the GDP benefit calculated from lower electricity costs in the previous post, the additional lower oil and natural gas prices are assumed to add an additional (pre-multiplier) $498Bn + $41Bn + $23Bn = $592.1Bn. At different multiplier effects (see previous post for a discussion):

At a MPC of 0.94, additional $9368 Bn of GDP
At a MPC of 0.5, additional 1242Bn of GDP

Combined with the previous gain in GDP from lower electricity costs:

At a MPC of 0.94, total GDP gain of 122.7% (more than twice as large of a GDP with zero cost electricity)
At a MPC of 0.5, a total GDP gain of 58.53%

What Does of This Discussion of low cost electricity have to do with stocks?

Well, this discussion is somewhat removed from stocks but has been inspired by a thought experiment of why the economy is not performing as expected currently with lower consumer expenditures on the horizon, and a shaky recovery (the recent Federal Reserve minutes demonstrated significant concern on the economy, and employment is disappointing). If electricity was available at a very low cost, likely the economy would be doing significantly better now and the stock market (as an index, with perhaps different/additional firms in it) would be much higher. As many economists have stated, GDP over the long term is determined by productivity, and productivity would be enhanced by low cost electricity -- and a higher GDP means that we would all be wealthier.

Wednesday, July 13, 2011

What impact would very low cost electricity have on the economy?

An interesting question to ask is what effect low cost electric power would have on the modern economy. In the United States, electricity rates are climbing upward, averaging 9.7 cents per kilowatt hour in the year ended February 2011, up from approximately 5.5 cents per kilowatt hour in 2000. (non-inflation adjusted, however) This is approximately a growth rate of 5.2% per year, a growth rate significantly higher than the average real GDP growth rate of the US of 1.5% per year over the same period.

Electric power is of course a prerequisite of almost every aspect of modern life, from the production and shipping of commodities (crushing reservoir rock, drilling into rock, pulverizing grains, etc) to manufacturing, to the operation of consumer products that rely on computing, lighting and/or moving parts or chemicals of any type.

The increasing cost of electric power in the United States is been driven to a large degree by higher coal and natural gas prices, which, combined, accounted for approximately 70% of electricity generation. One can certainly say that supply is failing to keep up with demand -- the reasons for the decline in supply relative to demand of coal in particular is not fully understood to my knowledge, as the supply of coal appears, on the surface, to be abundant. (I will address this topic of supply and demand in more detail below).

Alvin M. Weinberg's The First Nuclear Era: The Life and Times of a Technological Fixer captured the optimism of future low cost power -- nuclear power -- in the early 1950's until the early 1970's. Alvin Weinberg, along with the Nobel Laureate Eugene Wigner, were two scientists at the forefront of the development of nuclear power, and both theorized that fission and later fusion power would become exponentially more efficient (an electricity production version of Moore's Law) over the next few decades. As such, Weinberg describes (somewhat infamously) that nuclear power had the potential to be "too cheap to meter" which earned him and the nuclear community a significant dose of ridicule with the higher build costs experienced in the US in the 1970's.

Interestingly the potential of nuclear power was shared by other members of the scientific community in the 1950's and 1960's. One of the greatest scientists of the 20th century, John Von Neumann, wrote in 1955 that nuclear power was likely to be the source of vast amounts of energy by the late 20th century: "A few decades hence (nuclear) energy may well be free -- just like the unmetered air -- with coal and oil used mainly as raw materials for chemical synthesis, to which, experience has shown, their properties are best suited." Science Fiction movies such as "The Day the Earth Stood Still" (1951)implied that rapid technological progress in nuclear energy production was imminent (the movie showed Aliens coming to Earth due to the potential technological breakthrough from nuclear energy).

In 2011 however we are still stuck with a significant reliance on oil and coal for energy while nuclear is fading as an energy source, and energy is anything but free as the cost of energy is rising.

Estimating the Impact of Higher Energy Prices on GDP:

The most direct way to estimate higher energy prices is through treating higher energy costs as a "tax" on consumption on the residential electricity usage side and as a cost to business on the industrial and commercial usage side. Utilizing the GDP formula: GDP = Consumption + G + (X-M) + I, consumption should be reduced by the amount of the additional electricity costs on residential customers. Note that residential customers comprise approximately one third of US electricity demand, while industry comprises one-third of electrical demand and commercial (office buildings, retail, grocery, non-industrial) comprises one-third of demand.

Higher electricity costs on industrial and commercial sources will be treated as a direct tax to income in the income measure of GNI -- as GNI must approximately equal GDP then a simplifying assumption is that the higher rates pass through to consumption as well on the consumption measure of GDP. (actually I am not sure if this assumption holds, but will go with this anyway :)

The Keynesian Multiplier!

I've been informed that a higher amount of consumption would ripple through the economy through Keynes' multiplier. This is to say, that as a typical consumer has more income, they spent a percentage of this income, then the recipient of this income spends the income, etc, leading to a "ripple" effect through the economy. Keynes termed this effect "the multiplier" which actually (I believe) wasn't Keynes' original idea but one of his student's ideas (the Polish economist Michal Kalecki had a similar idea of revolving around additional stimulus impacting the economy, but did not come up with the idea of the multiplier).

With additional income that is present over all time periods -- and not during a recession (when consumers are more likely to save income to reduce debts and plan for an uncertain future) the marginal propensity to consume should be higher, towards the normal spending range for Americans, which is currently around 0.94. In times of recession, tax cuts tend to have a lower MPC, this paper argues that the MPC in the 2001 tax cuts was somewhere between 0.6 and 0.4.

The impact will be calculated according to the simplified MPC impact on GDP formula: Change in GDP = 1/(1-MPC)

So, a calculating the increase in GDP from electricity costs at a variety of rates per KWh verses the average rate of 9.7 cents in 2011:

Total Electricity Usage (thousand MegaWh) (2009): 3950331

Cost per KWh (cents) 9.7 8 7 6 5 4 3 2 1 0
Total Cost $383,182,107,000.00 $316,026,480,000.00 $276,523,170,000.00 $237,019,860,000.00 $197,516,550,000.00 $158,013,240,000.00 $118,509,930,000.00 $79,006,620,000.00 $39,503,310,000.00 $0.00
$ Savings (verses $0.097) $0.00 $67,155,627,000.00 $106,658,937,000.00 $146,162,247,000.00 $185,665,557,000.00 $225,168,867,000.00 $264,672,177,000.00 $304,175,487,000.00 $343,678,797,000.00 $383,182,107,000.00
GDP 2009 $12,832,600,000,000.00

Percentage Increase in GDP 0.52% 0.83% 1.14% 1.45% 1.75% 2.06% 2.37% 2.68% 2.99%

With a Marginal Propensity to Consumer of 0.94 for additional consumption from additional income, increase in GDP:8.72% 13.85% 18.98% 24.11% 29.24% 34.37% 39.51% 44.64% 49.77%

With a Marginal Propensity to Consumer of 0.5: 1.05% 1.66% 2.28% 2.89% 3.51% 4.12% 4.74% 5.36% 5.97%

Results: At first glance, with electricity at ZERO cost, the increase in GDP is appears to be only 2.99% from current levels, as a total, not in growth terms. (meaning that the savings from electricity are equivalent to about one year of additional GDP growth, as the US economy can easily grow about 3% in real terms during a good year).

However this result does not take into account the effects of the multiplier. With a normal multiplier of 0.94 -- meaning that consumers spend 94% and save 6% of their income -- the increase in GDP with free energy is nearly 50%. With a multiplier more in line with stimulus/tax break spending of 0.5, the increase in GDP from zero cost energy is only about 6%.

The 50% increase in GDP sounds more in line with what one should expect from zero cost energy, in a very unscientific way, mainly by being be more in line with the optimistic comments from the great scientists of the 1950's, referenced above:). Nevertheless, it is likely that the increase in GDP from very low cost energy would be even higher, for several reasons.

Reasons for likely higher GDP from Low electricity costs than implied by the multiplier:

First of all, low cost energy would likely translate to lower cost commodity prices across all sectors, in so far that cheap and abundant energy can be used to produce and extract commodities at a lower cost, leading to higher returns for oil, mining and agricultural firms. For oil refining, for example, electricity costs are approximately 10% of total costs. Lower costs would encourage higher supply, lowering prices relative to demand.

Low Cost Electrical Power as a Signal to the Market That an Energy Shortage is Not Foreseeable:

One aspect of very low electricity costs is that it would signal that a shortage in other energy commodities -- oil, natural gas and coal -- is not imminent over the medium term. There has been a lot of speculation on why the prices of commodities have risen over the past decade, for example this discussion forum at the Economist Magazine presents a wide variety of reasons for rising commodity prices, from Federal Reserve monetary policies, to demand from China. Financial speculation has also been blamed in some circles. One reason for higher prices is that an impending shortage may be anticipated relative to supply.

Adam Smith once asked, "Why is it that diamonds, although frivolous, are expensive, while water, while necessary for life, is free?" This question led decades later to the concept of supply and demand (invented by the economist Alfred Marshall, in (incredibly for such a core concept) the late 1800's. One can say, if water was in shortage, it certainly would be expensive -- the last remaining volume of water would be "bid" upon by millions of individuals (one can think of the movie "Mad Max" -- absolute chaos if water was in shortage relative to demand (although in that movie it was gasoline that was in shortage)(perhaps this mental picture is a bit silly but does illustrate the basic concept :). There is a lot of evidence that many key commodities, from oil to copper to even grains, have been getting close to shortage levels (the subject of this is lengthy, and for another post). For example, the additional supply of iron ore in late 2009 were bid upon by the European Union and China, with the only additional supply coming from three companies (BHP, Rio Tinto and Vale), causing the price to increase from around $80 per tonne to over $200 per ton. China, in particular, in late 2009 was desperate for iron ore as a key input into in construction-heavy stimulus program passed in 2008 -- the economic livelihood of China was at stake in obtaining iron ore as as such demand outstripped supply, causing prices to skyrocket.

Figure 1: The Historical Price of Iron Ore:

I would argue that a proper calculation of GDP savings would calculate additional price declines across energy as well as metal and food related commodities, stemming from low cost electricity. As this post is getting lengthy, I will calculate this GDP increase in the next post.