And the Oilsands are not alone; resource developments are in
play all over. The Québec government has put out a long-term strategy («le Plan
Nord») centered on resource development (esp. Mining and Hydro) in the
province’s North.  Republicans in the US relentlessly
continue to promote domestic oil production. Mining exploration and development
is active all over all over the globe. All this looks like a logical response
to the boom in natural resource prices that started before the last recession
and, with growing demands from industrializing nations in Asia and elsewhere,
is expected to continue.

So let me try to argue that speeding up resource production
in response could be a very bad idea. Let me try to argue that we should think
hard about doing exactly the opposite, even if it lowers GDP growth, because delay
may make us wealthier.

Just to be clear, I’m going to ignore any possible arguments
about environmental impacts, native land claims or social issues. I’ll ignore
market failures and labour shortages, transport bottlenecks, etc. for now (I’ll
come back to that at the end) so I can focus narrowly on the economics of
resource extraction in a simple world with perfect markets. I’ll use the
Oilsands as an example, but the argument applies generally to non-renewable
resources: oil, gas, mining, forestry (in Canada, we don’t farm trees – we mine
them) and forget about hydro power and agriculture for today. I also want to
focus on resource extraction:
exploration that adds to our known stock of reserves seems like good thing, as
do new technologies that make extraction more efficient.

For argument’s sake, suppose we have a pipe in the ground
that spews oil and we have a valve that controls the flow rate. (If you prefer,
you can suppose we have gold bars buried in our basement that we can dig up as
wanted.) The more we open the valve, the more oil we produce, the more we
increase real GDP.  Is opening that valve
wide a good idea? Could it be a very bad idea?

I think it depends on our wealth. In this case our wealth depends on the profit stream that
our well can produce. Because we’ve got a non-renewable resource, opening the
valve wider means we’re producing more barrels of oil today, but fewer in
future. It seems to me that we really want to be asking the question “Would we
prefer to pump oil today or tomorrow?”

The advantage of producing oil now is that we can earn
interest (i) on the revenue it generates. However, it also means that we get
today’s price of oil for our barrel instead of tomorrow’s. If that oil price is
expected to grow at rate g (g<>0), then delaying production by one period
costs us roughly i-g. The higher the interest rate, the more delaying our oil
revenues is costly. The higher the expected growth rate of oil prices, the
better it is to wait and pump our oil in the future to get a higher price for
it. So the key question should be, is i<>g?

At this point, many economists invoke Hoteling’s Rule.
Harold Hoteling argued back in the 1930s that non-renewable resources should be
produced at the rate where their expected price increase g equals the rate of
interest i. That makes a lot of sense
if you’re looking at the world market and you want to understand how the market
should clear. That’s not the question
I’m asking. I’m assuming that Canada is small relative to world markets; we
take the market price as given. (Probably a bad assumption for natural gas, better
for oil and good for gold.) For that matter, I think it’s probably a bad idea
to assume that all countries face the same i
and g; interest rates have not
moved similarly across major economies in recent years, and expected changes in
real exchange rates mean that g isn’t necessarily the same. So, as I said
above, I think the key question should be, is i<>g?

I’m not sure what the answer to that question is. Consider
oil: if I’m looking long-term, the
govt.’s 2010 survey of long-term oil-price forecasts showed the average
outlook was for just over 1% growth, with more bullish forecasts going up to
2.3% and bearish ones around -1.2% (all in real terms.) Meanwhile, the Bank of
Canada tells me that the yield on long-term government real-return bonds is about
0.5%. On that basis, some could argue that we should shut the value for now. I’m
not sure that’s right because I think the right interest rate should be higher
than that. Our long-run forecasts of oil prices are very uncertain, so I think
we need to add a risk-premium to that govt. bond yield.

And that’s where I think the debate the needs to be. How
bullish are we about the long-term prospects for oil (or gold, etc.) prices?
How uncertain are we? What kind of premium should we be using to adjust for
that risk? The more bullish you are on oil for the long-term, the more likely
that you should want to delay production. The more uncertain or risk-averse you
are, the more you should want to produce immediately. That same logic applies
to gold mines, forestry developments and any other non-renewable resource.

Has anyone heard this debate? Me neither. I don’t think
we’ve had to think seriously about this trade-off for a long time because it
seemed pretty clear that i>g. But both seem to have shifted in recent years.
Most have raised their estimates of g as demand pressures from industrializing
BRICs made themselves felt. Just after this took place, the financial crisis
caused a profound shift in real interest rates, dropping rates in favoured
countries (including Canada) to new lows. The result, as I think I’m made
clear, is that there can be doubt was to whether we should favour current
production or prefer to “save” our resources for a better day.

Now, to make my argument simple, I’ve ignored some
real-world factors that may complicate things. Perhaps most importantly, I’ve
assumed that anything we don’t produce today we can produce tomorrow; the
implicit trade-off is 1-for-1. If it isn’t, that skews our decision. If we
think current production is less efficient than future production (due to
supply bottlenecks or expected technology improvements), that makes delaying or
smoothing production more attractive. If we instead expect that the resource
will decay if we wait (think of trees killed by pine beetles), we may prefer to
produce now. I’ve also ignored demand elasticity by assuming that g isn’t affected by our production
choices. Where it is (Alberta heavy oil might be this year’s poster child for
this problem), that gives an additional incentive to shift production over time
to avoid glutting the market.

At this point, someone is bound to argue that a competitive
and free market will take into account all these factors and guide us to the
wealth-maximizing solution. I can think of three factors that might prevent
that.

1. If investors are overly concerned with
   short-term corporate performance rather than long-term wealth creation,
   managers will respond with too much production in the short-term.
2. If the supply bottlenecks or demand elasticity
   factors I mentioned above are important, they operate at the industry level.  Perfectly competitive firms will ignore such factors
   as beyond their control, causing too much production in the short-term.
3. Our politicians understand that we don’t vote on
   the basis of wealth creation; we’re more easily swayed by GDP growth and
   employment growth. I think Alberta politicians who are bullish on oil long-term
   and would therefore advocate a “go-slow” approach to the resource sector would be
   buried at the polls. I think much the same would happen in pretty much any
   democracy I can think of.  That gives politicians
   a strong incentive to encourage short-term expansion rather than long-term
   wealth creation. Usually that’s not a big difference; non-renewable resources
   may be an exception. 

This post was written by HEC-Montréal professor Simon van Norden