Where Does Oil Bottom?

By in
Where Does Oil Bottom?

Written the evening of 3/5/2020:

Outside of the Treasury market, oil prices have been one of the more reliable coalmine canaries when it comes to equity market returns in recent years:

  • West Texas Intermediate prices swooned 31% in October/November 2018, even as the S&P 500 only dropped 5%. The pain would come for equities in December, down 10%.
  • WTI caught a bid, along with equities, in Q4 2019. But it dropped 17% from January 1st to February 19th 2020, even as the S&P was hitting new highs.
  • Today’s close of $46.01/barrel (down 74 cents) is uncomfortably close to both last Friday’s 1-year lows ($44.76) and the December 27th, 2019 lows ($44.48).

That raises the question, “where do oil prices finally bottom, and can we use that as a bullish signal for US stocks?” To set the stage, here is a long run price chart for WTI from 1986 through January 2020 so we can see where oil prices show their most notable downward moves:

#1: From that chart we can pick off the prices where oil showed sharp declines and see what the S&P 500 did in the 12 months after those lows:

  • December 1998: $11.35, due to the Asia Crisis
    Inflation adjusted to today’s dollars: $18.00
    S&P price return next 12 months: +27%
  • December 2001: $19.39, in the economic aftermath of 9-11
    Inflation adjusted to today: $28.14
    S&P price return next 12 months: -23%
  • February 2009: $39.08
    Inflation adjusted: $47.00 (note: essentially same level as today)
    Forward 1-year return on the S&P 500: +50.3%
  • February 2016: $30.32
    Inflation adjusted to today’s dollars: $32.59
    Forward 1-year return on the S&P 500: +22.3%

The upshot: using the most recent inflation-adjusted low on oil prices ($33 in 2016), we’re still far from WTI flashing an equity buy signal since it trades for $46/barrel today. While the prior periods are historically interesting, there were different supply/demand fundamentals in place and therefore we put less weight on those observations.

#2 & #3: We also want to review how oil prices compare to both stock and gold prices on an absolute basis. This is a bit of an unconventional approach, but here is our thinking:

  • All three are priced solely in dollars and have robust long time series (back to the 1970s, courtesy of the World Bank).
  • Gold and oil are both commodities but capture different fundamentals, with gold being tied to confidence in US monetary policy and oil linked to worldwide economic growth.
  • Stock prices over the long run reflect corporate earnings power, driven by human ingenuity. But, over the short-medium term they can get overly bullish/bearish relative to oil, which while subject to innovation in terms of supply remains the same in terms of uses (and the world still consumes close to 90 million barrels/day).

There are historical charts below for S&P 500/oil (i.e. how many barrels it would take to “buy” one S&P unit) and gold/crude (how many barrels to “buy” an ounce of gold); here is an analysis of where these trade just now versus long run norms (note: oil prices quoted are an average of WTI and Brent for comparability to the World Bank data):

S&P 500/oil prices: 62.9x

  • This is well above the 10-year average of 31.0x and right on the 2 standard deviation upper bound (63x).
  • Worth noting: the averages over the last 20 years (30x) and 30 years (33x) aren’t far from the 10-year average, and nor are 20-year (15 points) and 30-year (19 points) standard deviations.

Bottom line here: we would argue that S&P 500 valuations are historically higher because of low interest rates, but being 2 standard deviations from the mean says either oil is cheap or stocks are expensive. Or, of course, both.

Gold/oil prices: 34.8x

  • This ratio is also well above its 10-year average (19x), as well as 20-year (15x) and 30-year (17x) mean…
  • … And it is higher than the 2 standard deviation upper bound over the last 10 (31x), 20 (27x) and 30 years (29x).
  • Using just the last 10 years of data, we hit the 3 standard deviation upper bound at 37x, implying a low print on WTI of $43/barrel, 6% lower than today.

Bottom line here: as much as we totally understand why gold is ripping (emergency Fed rate cuts, more expected) and oil is under pressure (COVID-19 economic shock), it is important to understand that we’re nearing a 3 standard deviation upper boundary. That reminds us that when markets get truly wobbly, gold tends to decline over the short term. In 2008, for example, it was down 6.4% on the year.

Final thought: oil prices are threatening to break to new lows even though they already are notionally cheap to their long run relationships with stocks and gold, and any significant decline from here (more than $1-$2/barrel) likely signals more financial market volatility.