The Hyperinflation Hoax is worse than we thought: We can’t change the price of Oil
Remember that post about the futility of using monetary policy to impact the price of oil? It turns out that the elasticity of oil is even lower than I thought! The IMF – ok, no jokes – came out with a report showing the short term price elasticity is nearly irrelevant, while the long term price elasticity is freakishly tiny. The Blog world is a-titter about this.
The IMF says that a 10% increase in the price of oil results in 0.007% less demand in the short term.
The U.S. uses about 15 million barrels a day. A 10% increase in the price of oil results in – wait for it – a decrease of 1050 barrels of oil per day. Increasing the price of oil from $90 to $99 results in 1000 less barrels of oil demanded. That is 1 futures contract at the CME/NYMEX. A 1 lot. To put this in perspective, 319,000 contracts traded yesterday.
What level of interest rates – in the standard model- could be high enough to push down oil prices 20% or so, back to$90 a barrel? The demand for oil will not change in the slightest even if rates were increased to 5% tomorrow. People won’t say “Oh, this quarter point increase in the Fed Funds rate makes me want to use less gasoline – or at least pay a lower price for it.” A 10% increase in the actual price impacts demand by an amount so small it is a rounding error. Tiny movements in interest rates will have zero observable impact on the price we are willing to pay for gasoline.
A self-inflicted cure of higher rates in response to high oil prices would be far worse than the disease. We know what it takes to get the price of oil down to $40 a barrel. It takes a global depression. It takes losing 500,000+ jobs a month in the U.S.
This is another reason to fight the Hyperinflation Hoax.
[Update: I made a mistake in oil demand. The actual numbers are a 10% increase in the price of oil causes a drop of .2% in demand after 20 years.
This translates to a decrease of demand of about 30,000 barrels a day. It does take 20 years for this to happen. So if we assume a linear drop, then it's a drop of about 125 barrels/day a month, every month for 20 years. This is still very small. I wonder how this number is significantly different than zero - it seems impossible that such a small effect would have an error term that's even smaller, considering the data used. ]