« An academic integrity policy for faculty? | Main | British Tuition Hikes: A Canadian Opportunity? »


Feed You can follow this conversation by subscribing to the comment feed for this post.

The above post was by Simon van Norden (just to beat Stephen to it!).

Neat Simon. Very simple. Funny we don't hear much about Norway. Scott Sumner has done a couple of posts talking about Sweden, specifically its monetary policy. I wonder what Norway's monetary policy has been? (I'm totally ignorant on that topic.)

I'm not sure this comparison is makes much sense. Unless I'm mistaken, oil production in Texas (per capita) is an order of magnitude lower than in Alberta and Norway. Texas is not really an oil-dominated state, it's a state where the oil industry is considerably bigger than the US average. It has a highly diverse economy, which continues to grow very fast during years when oil prices are low. It also grows many times faster than neighboring states like Louisiana and Oklahoma, which have as much or more oil per capita.

Scott: what about Norway's monetary policy? Any thoughts?

(I'm pissed off, because while Scott S. is off writing really interesting posts, I'm stuck here explaining S=I to MMT followers, because the real MMT leaders don't seem to be keeping their followers straight on this issue.)

OK. Norway targets 2.5% CPI inflation. But it ignores the effect on CPI of interest rates, atxes, and temporary disturbances, so it's really targeting some sort of core inflation. It says it is a "flexible inflation targeter" giving weight to both inflation volatility and output/employment volatility. Inflation is currently below target. The policy rate never went down to the ZLB. It has been rising recently, and is currently at 2.25%.


Scott: I'm no expert, but if Texas is a more diversified economy (i.e. with lots of refining, oil services, corporate headquarters, etc.) wouldn't they are much better positioned to benefit when energy prices rise than their neighbouring oil-rich states?

Why don't you bring some statistics to the discussion to support your claims? or at least suggest some sources?

I'd be particularly interested to see how very fast Texas continued to grow in the late 80s when oil prices were low. And when you say "many times faster", I assume you mean more than 2x, right? ;-)

Nick: I know more about Norwegian monetary policy than about the Texas economy, but that doesn't say much! Their monetary policy mostly worries about the "on-shore" economy (i.e. non-oil production.) They had the good sense to not to adopt the euro (with volatile oil prices, floating exchange rates can come in very handy.) But mostly I'm impressed with the cutting-edge work their central bank does on density forecasting and transparency. The Norges Bank regularly publishes their forecast distribution of interest rates, among other variables. Not many other central banks come close!

According to the BLS, oil and gas extraction accounts for 154,000 jobs in the entire US. Petroleum and coal manufacturing gets you another 111,960. Pipeline transportation is good for another 42,000. Job growth in those sectors alone can't account for all of the job growth in Texas - even if you assume that all of those jobs are in Texas in the first place.

What is more, the employment trends the two other big oil states (which are less diversified than Texas) have been less impressive (Dec 2007 employment/current employment):
Louisiana: 1.97 M/1.88 M (-4.5%)
Alaska: 331.7 K/337.2K (+1.6%)

Oil is part of the story, no doubt, but it would be fallacious to assume it is the whole thing.

Simon: yes, I was impressed by their "fan charts"! (I'm against CB's publishing interest rate forecasts, BTW; I think it puts too much focus on the *perceived* instrument, rather than on the target variable, where it belongs. But that's a debate for another day).

Simon: If you're looking for stats a quick search reveals the following numbers:

Texas produced 365m barrels of crude oil in the past year. Population is 25m, so about 14.5 barrels/person/year.

Norway produced 650m barrels of crude oil in the past year. Population is 5m, so about 130 barrels/person/year. 

Alberta produced 550m barrels of crude oil in 2009. Population is 3.7m, so about 150 barrels/person/year. 

So Alberta and Norway produce about 10 times as much per capita as Texas. And refining is irrelevant.  Refining profits (the "crack spread") are independent or even negatively correlated with the price of oil. And "diversification" is just another way to say "produces less oil".  Obviously not helpful when looking to gain from a rise in the price of oil.  

So Scott's right: There's no justification for a direct comparison of employment growth in those economies. 

The earlier figures I posted seem to be weird. The BEA says there are 799,100 oil and gas extraction jobs and 117,200 in petroleum and coal manufacturing. Between 2007 and 2009 O&G saw the fastest employment growth of any industry, increasing by 92%. Number 2 was securities investment, while the auto sector fared the worst. So maybe oil is a big part of the Texas story.

Didn't Matthew Shapiro already tote up how many Texans were in extraction and everybody linked to that post?
Maybe I don't have an inclusive enough definition of "everybody".

Um, that graph looks like it says that Texas has had 0.5% growth (and I'm being generous with rounding) over the space of a year. They still haven't regained their total pre-recession employment and aren't coming anywhere close to that for another six years as an eyeball estimate.

Why is Texas crowing again?

BTW what was their population growth in working age people? At this rate they might not have even kept pace with population growth.

Politicians - they always take credit for growth, and always pass the blame for contractions.

To change topic, that is a sexy graph!

You should think about it a bit differently:

365 million of barrels of oil produced last year was worth about 35 billion dollars, or about 3% of Texas GDP.

A 3% GDP stimulus that in significant part flows back into the local economy (possibly more, if it has a multiplier higher than 1.0x) is a great way to soften a recession by bridging the output gap and keep growing ...

There's a limit to such stimulus when you are near full employment: Alberta (around 5%) and Norway (around 4%).

So the Texas story reminds us that a significant keynesian stimulus works very well in recessions - regardless of what the source of the money for spending is: government spending or being a lucky state that can simply pump oil out of the ground.

Reading the various posts this morning, we can probably agree on two things.
1) In recent years, having oil helps.
2) How much it should help is debatable.
So let's talk about (2) and see what we can agree on. How about the following five statements?

A) Excluding oil extraction doesn't properly capture the impact of having oil. (This goes doubly for employment, since a working oil well directly employs very few people.) An integrated oil company may employ far more people in HQ, exploration, transport, support services, etc., than in extraction. And we haven't begun to consider the multiplier effects of their direct employees' demand for housing, food, entertainment, legal services, financial services, etc. I can't think of anyone who would come out and defend this way of measuring the benefit of a rent-extracting industry.

B) During times of high prices, having exportable natural resources means that you collect economic rents. I think of this as being like a helicopter drop of wealth, not money. (Those of you who did international trade: just think "transfer problem".) The question is, how much of the wealth stays where it is dropped? That's where a diversified economy comes in. A vertically integrated economy traps more of the wealth by ensuring that more of the value added is produced locally. (This should not be a new idea to Canadians; that's why it was illegal to export logs for about a century.)

C) I'm no expert (and hopefully an expert will step in here with some hard facts), but I would have guessed that Texas exports lots of oil-sector materials and expertise outside the state. (Think Brown & Root, or Boots & Coots.) That would mean that they derive more benefit from high oil prices than oil-production/person figures would suggest. Probably more than, say, Alaska or Louisiana.

D) Now, if you want to say that good government or low taxes or Christian values or generous tax breaks caused Texas to have a diversified economy that exports goods and services all over the world, that's an interesting argument. But notice that (1) that diversification was way before Perry's time, and (2) that diversification into oil-related exports still means their economy is unusually sensitive to oil prices (but not so much to the amount of oil extracted in Texas.)

E) A bearded economist from Albany recently won a Nobel Prize for his models explaining why some specialized industries concentrate geographically and then export their expertise widely. Sounds like the kind of thing that might help to understand what is happening in Texas.

So step back and look at the big picture: 6% employment growth over 5 years is proudly cited as evidence that a candidate is qualified to become president.

Who was is that said "In a democracy, people get the government that they deserve....good and hard!" Heinlein?

Not Heinlein but Mencken.

Determinant, it's really the population growth that is extraordinary in Texas and we want to explain. Ed Glaeser attributes it primarily to the lack of regulation on real estate development, giving rise to cheap housing prices.

....not to mention having a long border with Mexico?

So I wondered about the view of Texas as a highly-diversified economy that is not much affected by oil prices. I went looking for an opposing view and quickly found it: The Texas Alliance of Energy Producers (http://www.texasalliance.org/) Not surprisingly, their "research foundation" has a different view. (http://www.foundationforenergyeducation.org/) Their latest education pamphlets (http://www.texasalliance.org/admin/assets/A_Joint_Association_Education_Message_Protected_PDF.pdf) have a handy "Oil and Gas by the Numbers" section with lots of claims

- "In fiscal 2009, oil and gas companies paid an average of $27,000 per employee in state and local taxes and royalties. By comparison, other private sector companies averaged only $4,800 per job. This five-fold difference highlights the critical role oil and gas plays to keep the Texas economy - and Texas government - viable and resilient
- "According to one economic model, the highest job multiplier in Texas is in petrochemical manufacturing, at nearly 17 additional jobs created for every oil and gas job. The second highest multiplier in Texas is in petroleum refining, creating 14 additional jobs for every refining job."
- "Oil and Gas jobs in Texas: 315,000. Average Oil and Gas Wage: $107,289"

Okay, that only proves that the Energy Industry wants you to think that it is Vital to Texas. My point is just that the multiplier effect needs to be taken seriously.

I also learned something on their site; there's a big investment spending boom in Texas in recent years in shale gas exploration.

The topic of moving to Texas come up frequently at work. Its the only state in the union ever discussed regularly. The conversation tends to run like this:

House prices are cheap down there and there is no income tax.
Yeah but it's hot.
Plenty of money to run your a/c after those mortgage payments.
But what about all those conservatives.
Austin is almost entirely democrat.

Oh okay. I should check it out.

When people say a place has an oil economy they are thinking about an extractive enterprise. That's not what Texas has. Texas is the skills center of the oil industry. These are highly technical jobs not your typical make work at the oil ministry or state coffers filled with black gold.

I think that makes the suggestion here prejudicial.

"So step back and look at the big picture: 6% employment growth over 5 years is proudly cited as evidence that a candidate is qualified to become president."


Surely Texas trades more with the other lower 48 states than does Alberta. And surely its affected more by the Fed policy than is Alberta. So if the Fed is running a sub-optimal monetray policy which is depressing demand in the US, surely this will have a much more negative impact on employmnet in Texas than employment in Alberta. So I don't see how showing a chart of Texas vs. Albeta employment tells you anything at all.

The question you really want to answer is: has Texas done better (in trems of real output and employment) than other US States after controlling for the effcet of the relative improvement in its terms of trade?

"investment spending boom in Texas in recent years in shale gas exploration"

Yup. That's the story. Texas oil production is in decline and has been for many years. A 1 second google search give this graph:


These days it's all about the shale gas. Shale gas is tight - meaning it requires lots of drilling (horizontally through the rock, not just straight down) and fracking (they pump a slurry of acid and sand under extremely high pressure and thus fracture the rock and open up channels for the gas to flow). It requires all sorts of specialized equipment and labor.

Also, the demographics of the oil industry in Texas are just awful. The industry went through a terrible depression in the 80's and 90's. Nobody was hiring or training new people. So now they are faced with the mass retirement of all the workers who make the industry go, so they are desperately trying to find replacements. Another one second google search gives these (somewhat old) articles:


Jon: Your characterization of Texas as the "skills center" of the oil industry sounds reasonable to me. That should make much more sensitive to the price of oil than the oil produced per capita measures suggest. It is also consistent with some describing it as a "diversified" economy. Do you and I agree on that?

G. Bush: We agree that Texas is much more exposed to US monetary policy than Alberta (or Norway.) We also agreed on what the real question is. (Neither my post nor Gov. Perry's claims are completely reliable answers to that question.) On my planet, however, US monetary policy is more stimulative than Canadian; that counts against Texas.

Patrick: Thanks for the facts! (I love people who bring facts and reason to the table.) Your description of the late 80s and 90s jives with my vague recollections; so what do you make of Scott Sumner's claim that Texas "...has a highly diverse economy, which continues to grow very fast during years when oil prices are low."?

Thanks for the data K, I thought it was roughly an order of magnitude different, but didn't no how close it was to being one tenth as oil intensive as the other two. Not a bad guesstimate!

I was thinking of the long term growth in Texas, I agree with commenters who point out that the recent rise in oil prices has something to do with the rate of growth in recent years. But recall that Texans use way more gasoline than people in places like NYC. So the high oil prices badly hurt the 95% of Texans not in the oil industry.

Nick, I know nothing about Norway monetary policy. I'm struggling with MMTers to. Now one says a currency can't be devalued without first boosting demand. Is that MMT theory?


"On my planet, however, US monetary policy is more stimulative than Canadian; that counts against Texas."

Really? By what measure? I don't want to get into an argument about what the best way to judge the stance of monetray policy is (or do I?) but surely AD is much lower relative to potential in the US than in Canada or Norway. And that hurts Texas more than Alberta or Norway. So comparisons with Alberta and Norway that don't take this into account are highly misleading.

Claim: "... has a highly diverse economy, which continues to grow very fast during years when oil prices are low"

That's an empirical question, isn't it? So I went here:


And charted nominal GDP, real GDP, total employment, and a bunch of others for Texas, California, and New York for no other reason than they're all fairly big states. Long run, they all look pretty similar on most measures (at least the shape of the graphs, if not the levels). Total employment for California and Texas look very, very similar (Texas is just shifted down). No doubt an industrious data nerd could download the tables and normalize for population. That's not me.

The whole US economy is a catastrophe. Are the policy elite really at the point where the best they can do is argue over random wiggles in state employment data to lay claim to being the least bad?

To quote Nick: "Oh Christ"

Gregor Bush: On my planet, the Bank of Canada Target Overnight Rate is higher than the Fed Funds rate, it is not at its minimum (while the Fed Funds rate is), the Bank of Canada has not pledged to keep its rate at the minimum until 2013 and it has not engaged in programs of Quantitative Easing, just to mention four measures. (And did you notice that the CAD is relatively strong against the USD?)

Sure the output gap is bigger in the US than in Canada; but why would that mean that US monetary policy is tighter rather than looser? (Think of a Taylor Rule, for example.)

Simon: there's more than one way to measure tightness/looseness of monetary policy. Even if one does think of monetary policy as setting a nominal interest rate, whether that interest rate is low or high can only be judged relative to what would be needed to keep the price level (or nominal GDP, or some such variable) growing at target. The Bank of Canada is (roughly speaking) on target. The Fed doesn't have a target, but if we impute any sort of implicit price level, inflation, or nominal GDP target to the Fed on the basis of past data, the Fed would seem to be below target. On that basis, you could say that monetary policy in the US is (or has been recently) tighter than in Canada.


On my planet monetary policy was much tighter in Japan in September of 1998 than it was in the US despite the fact that the BoJ target rate was 0.25% in Japan while the fed funds rate was 5.50% and the US dollar was at a very high level against the JPY. I cite as evidence for this the subsequent rise in (overall and core) inflation and in the US to above 3% by mid-2000 and the subsequent worsening of deflation in Japan over this same period. In 2002, when the BoJ started it second round of QE, monetary policy was still tighter than in the US.

I’m not crazy about the Taylor rule as an estimate of the stance of policy because it assumes a constant equilibrium real rate and it is too backward looking. But even by this measure (if you believe the BoC and CBO estimates of the output gaps) the Fed is currently tighter than the Bank of Canada. Note that I’m assuming the Fed is targeting 2% inflation, consistent with their long-term forecast:

Taylor Rule stance = policy rate – (2 + core inflation + 0.5*(inflation gap) + 0.5*(ygap))
Fed stance = 0.12% - (2+ 1.8 + 0.5*(-0.2)+0.5*(-6.9)) = -0.13%
BoC stance = 1.5% - (2+ 1.7 + 0.5*(-0.3)+0.5*(-0.7)) = -1.7%

If you were using a forward looking Taylor rule and you thought that the outlook was worse for the US than for Canada, then the Fed’s policy stance would look even tighter on a relative basis (for the record I think the Canadian output gap is larger than the BoCs 0.7%).

“The money rate of interest is, in reality, very often low when it seems to be high and high when it seems to be low.”
-Knut Wicksell

Sorry, the current overnight rate is 1.0%, not 1.5%. So the current BoC rate minus the Taylor Rule implied rate is -2.2%, not -1.7%.

Nick: You're measuring monetary policy relative to what is needed. I'm measuring monetary policy relative to historical and international norms. You're saying "US monetary was not stimulative." I'm saying "US monetary was very stimulative....but not as stimulative as they would have wanted."

Simon: but it's meaningless to judge the tightness of monetary policy without reference to the natural rate of interest. If Chinese potential growth is 10% and target inflation is 5% then the equilibrium rate is around 15%. By your measure that would be tight. But it isn't, it's neutral. If the BOC changes it's inflation target to 5% then a 5% policy rate will be inflationary, though historically it's about neutral. There is just no significance to the historical level of nominal rates.

Also, you cant call a policy "stimulative" if it's deflationary.

K The natural rate of interest is operationally meaningless; for example, see Clark and Kozicki (2005). We've seen the same with different monetary frameworks: growth rules that become a nonsense with the velocity shocks that we've seen of late in the US; MCIs that required highly variable "target" MCIs. And don't get me started on output gaps and Taylor rules....

The word "cant" obviously does not mean what you think it does: this is a logically consistent usage of the term "stimulative" as well as the common usage. (The same applies to Nick's comments that interest rates can "only" be judged relative to what is required.) Just look at the counter-intuitive spin you have to put on recent events in your framework; a negative fiscal policy shock implies that, overnight, without any fed action, monetary policy is suddenly much tighter. Using a conventional definition of "stimulative", you'd say monetary policy needs to become more stimulative to offset the negative fiscal shock. G. Bush's interpreation of BoJ policy in the 90s shows just how bizzare such a definition is.

But if you want to talk strictly about logical consistency, the real problem with solving the entire economy to decide whether you want to class monetary policy as tight or loose is that it ignores problems of monetary and fiscal policy co-ordination. If you apply the same logic to fiscal policy, then classifying fiscal policy as tight or loose will realistically require you to condition on monetary policy and vice versa. Seems to me that that's not helpful.

G. Bush:

Well, it looks like the argument about how to measure monetary policy is underway! But I want to come back to what I think you were trying to say about Texas.

Surely Texas trades more with the other lower 48 states than does Alberta....So if the Fed is running a sub-optimal monetary policy which is depressing demand in the US, surely this will have a much more negative impact on employment in Texas than employment in Alberta.

I think we agree that Texas has a larger exposure to a sluggish real US economy and a larger exposure to US monetary policy than Alberta. I think you're arguing that both are a drag on the Texas economy (which makes my comparison to non-US economies unfair.) The claim I made is that monetary policy is not a drag on the Texas economy because (on my planet) US monetary policy is "loose." But let me try to look at this from your (and K's and Nick's) perspective. We can't say whether Fed policy is tight or loose until we calculate the natural or "required" rate of interest for Texas. I really don't know what that might be, but I'm guessing that (from your perspective) Fed policy looks much looser for Texas than for the nation as a whole.

If I understand your approach to "tight/loose" policy, you need to work out the target or natural interest rates for both Alberta and Texas before you can start to say anything about the differential impact of monetary policy on their employment growth. Is that right?

Yes, I think that's right.

And there are two things that make the natural rate of interest much higher in Alberta than in Texas (and therefore monetary policy much looser in Alberta than in Texas despite the fact that short-term risk-free rates are about 90bps higher in Alberta). The first is that monetary policy is effectively looser in the rest of Canada than it is in the rest of the US. This helps boost demand for Alberta-produced goods and services relative to Texas-produced goods and services. The second factor is that the positive terms of trade effect that you mentioned is much stronger in Alberta than in Texas as Alberta’s net exports of energy products are much larger as a share of GDP. Even more importantly, its expected future net exports of energy products are projected to grow much more quickly than those in Texas which has driven the surge in investment spending in Alberta over the past six years.

So yes, if Alberta and Texas each had their own currency and a central bank with a 2% inflation target, I’m fairly certain that the short-term interest rate differential would have been wider than it was over the 2005-2010 period. That’s not to say that the economic policies of Texas and Alberta do not themselves affect the natural rate of interest, but I think the key drivers of the difference between what was observed and this hypothetical case are the two factors cited above.

Imagine if I had looked at a region of Japan in the year 2000 with an industrial structure that was heavily concentrated in the tech sector. And then I looked at output and employment growth in this region and found that it had badly underperformed Silicon Valley and Waterloo over the last four years. Would it have been a good idea for me to draw any conclusions from this about the economic policies of that particular Japanese region?

Keep in mind Canada had employment slack during after effects of Mulroney deficits. Be nice to see the actual employment chart instead of %. Probably the last 20 years of human capital investment need to be normalized for that chart to work.
The most useful metric to me during recession was the USA Bureau of Labour's (their Stats Crown) measure of applicants to help wanted ads. I think it was around 6/1 3 years ago. It would be great to have a metric that measures how long it takes until hired. Take the mean/mode/median of the number of applicants per advertised job, and measure how long it took until hired. This would tell where the economy is hyper competitive. Where not efficient, a Crown day labourer, and skilled such amnetities (forklift course and placement search paid for) could be enacted to bootstrap. Modifying my GAI position to two tiered. For all based on innate wealth; Auzzie's resource tax is a good proxy. And then a bigger personalized GAI based upon doing something as reasonably useful as one can.

..examples of differing human capital include: better healthcare in AB than TX. Bigger potential migration pool in TX than AB. Cheaper education in AB and Norway than TX.
Cities are different. With a temporal applicant/help-wanted-ads hired metric, you could put (esp N.Def) pork where efficient as stated. Could also do the P.Martin merit-based foreign aid thing and give to regions that improve collusion practises. There is a mininum time to consider applicants; maybe 1/2 wk for unskilled and two weeks for skilled positions, or something like that. Beyond this, start the clock.
I don't want to work for bosses who grew up under Stalinism or work with subpar coworkers. I don't want transfer payment supporting this. This metric would tell where to yank/reinforce transfer payments and pork. Transfer payments are okay where work and effort are cultured. If the payments themselves reinforce parasitic behaviour it is a tax.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad