Systemic Collapse: The Basics

[Crispy]

Right, ok, that makes sense. There is a hell of a lot of guesswork in H, though. I also don't see any particular justification for the path of D. If we have hit peak oil, necessity will lead to massive investment in alternative sources of energy. That 'D' may be a realistic demand for energy (and it very well may not be - a prolonged period of near-zero growth would see demand for energy not rising much at all), but that isn't the same thing as demand for oil.

D is demand for oil only, in this graph.
And oil has no effective replacement in the "alternative" category.

 

[Falcon]

you have not engaged one bit with the facts i've presented to you - and you're right there is no basis for discussion with someone who does this

Ironically, I even complimented you on them at one point. To the extent that you articulate the mechanism of US economic failure and the subprime mortgage crisis, your account is perfectly adequate. There is nothing to engage with you in that context since I agree with you.

It is simply the case that the global financial crises can, in my view, be explained better by the systematic rise in oil price which was observed immediately prior to it. Moreover, that systematic rise in oil price can be explained better by the supply contraction that was observed in turn immediately prior to that.

I'm not sure what you believe I have failed to engage you on. Bubbles create price rise. So does scarcity. I asked you to explain why you think the price rise was bubble not scarcity induced, but you refused. The US economy affects the global economy. So does the global economy affect the US economy. I asked you to explain why you think an effect taking place within 25% of the global economy is more plausible than one acting on the whole economy, but you refused. I've cited academic analysis of empirical evidence, but you've accused me of not supplying evidence, then accused it of being inadequate, then confessed you aren't familiar with it. And throughout you've been thoroughly unpleasant toward me in what ought to be an interesting exploration of a difficult and important subject.

 

[littlebabyjesus]

D is demand for oil only, in this graph.
And oil has no effective replacement in the "alternative" category.

Ok, well I still don't see justification for that massive rise in D. And it can't zoom ahead of supply like that. The factories/power stations/cars simply won't be made if there isn't supply there to make/run them.

Oil has no effective replacement at the moment, but if production starts to decline, that will lead to enormous efforts to find replacements. That's one of the continuing flaws with all these conjectures - they treat people as if they were passive in this.

 

[love detective]

It is simply the case that the global financial crises can, in my view, be explained better by the systematic rise in oil price which was observed immediately prior to it.

can you state in a few simple sentences what the empirical evidence is which shows the causal impacts of oil prices rises on the financial crisis? all you've done so far is to point to a tripling (although your graph shows a doubling) of commodity prices in a three year period. I've taken the numbers that this boils down to (provided by free spirit) and shown that these are a fraction of the impact caused by the unfolding of the subprime lending debacle. Please show me, clearly and concisely why these small impacts had a much large impact than things which had a much larger impact on household finances. You simply haven't done this yet, you've ignored my rebutals of your point and continued to assert your point. Can you do that now for me?

Make your case and point me towards the data which shows that these price inflation increases were the single biggest thing contributing to sub prime borrowers inability to keep up with their mortgage payments? If it's so obvious that this was the cause, the facts & figures must be readily available for you to provide to me now - so please, if you want me to take your theory seriously, back it up

 

[Crispy]

And it can't zoom ahead of supply like that. The factories/power stations/cars simply won't be made if there isn't supply there to make them.

That's precisely the problem that the graph intends to illustrate.

Oil has no effective replacement at the moment, but if production starts to decline, that will lead to enormous efforts to find replacements.

I agree. But can it be done fast enough? Previous energy source transitions have been done during times of continued good supply of the previous source, and the replacement has been of a higher quality. We must therefore pin our hopes on a literally miraculous breakthrough. Fingers crossed!

 

[littlebabyjesus]

That's precisely the problem that the graph intends to illustrate.

I don't see how. The divergence of the two lines is meaningless.

 

[Falcon]

Right, ok, that makes sense. There is a hell of a lot of guesswork in H, though.

H is one of an infinite family of curves constrained (amongst other factors) by the observation that the area under H cannot exceed the area U (in english: "we can't produce what we haven't discovered"). The one shown is an educated guess. Specifically, it is the curve you get when you extrapolate exponential depletion of resource base U under the assumption of constant commercial and technical constraint (in english: "business as usual").

To get a different outcome, you would have to have different commercial and/or technical conditions relative to historical baseline. So just spending what you spent before gets you H. To get more than H in any year, you have to spend more. We are spending less in real terms, by the way.

Investment can only do two things: find more oil, or accelerate previously found oil. We can reasonably estimate how much oil will be found. All we can do now is accelerate it. Some theoretical alternative (H) are shown here:

Acceleration comes at the cost of increased depletion rates post peak. If you could achieve perfect piston-like displacement of the global petroleum system (you can't) you would fully exhaust all discovered and foreseeably discoverable oil by 2030 and production in 2031 would be zero(ish).

I also don't see any particular justification for the path of D. If we have hit peak oil, necessity will lead to massive investment in alternative sources of energy. That 'D' may be a realistic demand for energy (and it very well may not be - a prolonged period of near-zero growth would see demand for energy not rising much at all), but that isn't the same thing as demand for oil.

In my graph, it is oil demand derived pro-rata from global energy demand projected at 1.45% per annum growth i.e. it does not account for substitution by other energy sources. (1.45% is what you get when you partition the global population into OECD and non-OECD groups and project each group's population growth and per-capita energy growth. Any other outcome is basically telling 80% of the planet they can't live half as well as us - good luck with that).

It does this on purpose because the resulting gap, specifically, (D-N), defines the quantity of energy capacity from all sources (conventional oil, non-conventional oil, nuclear and renewable) which must be commissioned in each year to sustain energy demand growth, given natural depletion. You can convert the energy equivalent of the gap in each year into kwH and from there into alternatives to estimate capacity growth schedules for different combinations of technology. Since all sources (including conventional oil) must compete for capital, materials, manpower and land, it delineates the aggregate resource demand schedule implied by sustaining growth. We don't have anything like the capital, manpower, materials or land.

There is also a structural problem there - demand for energy can never zoom off ahead of supply like that.

Quite. Hence the title of the thread: "Systemic Collapse: The Basics"

 

[elbows]

I think some people were hearing you talking about 10% decline and assuming that you meant actual total production would fall off that 10% cliff.

Thats why I've been moaning, because some of Falcons posts actively encouraged this error, often by attempting to describe the necessary new production that will provide some offset as somehow being unreal. Since I am not a fan of mainstream estimates which have tended to fudge the future production numbers to fit the expected demand a few decades down the road, I believe that at some point the new capacity is not going to be able to offset the decline, but my issue with Falcon is the impression he sometimes creates that we already reached that point. Rather I would expect that moment to only be solidly identified in hindsight, and so some of the woe timers he wants to start ticking now may not actually start ticking for some time to come.

 

[littlebabyjesus]

We don't have anything like the capital, manpower, materials or land.

Here I have to disagree. When the world finally faces up to both the energy crisis and the need to change how we use that energy, there will need to be an enormous, coordinated worldwide effort. When you look at the capital directed towards vanity projects such as the Apollo programme, and the continuing capital directed towards the military, you can see the kinds of efforts that can be made where there is the will to do it. There will be the will eventually, because there will be no alternative. At its peak, the Apollo programme had half a million people working on it. Make that kind of effort in the US and right around the world to find a sustainable alternative to oil and the only certainty will be that we will discover all kinds of things that we can't know about before we start looking.

 

[elbows]

Elbows, whithout nagging you like your mother, I can only keep inviting you to read that link I gave you to IEA World Energy Outlook 2008. It is well written and extremely informative.

Current global conventional production capacity is 80 million barrels a day. The natural (i.e. uninvested) depletion rate of the global petroleum system is 10% per annum (Source: IEA WEO 2008). The halving time at 10% depletion is (70/10=) 7 years. Half of 80 million is 40 million. The global production system will therefore lose 40 million barrels per day of production capacity in the next 7 years through natural depletion. (Yes, these apparently small depletion rates yield colossal losses in short time scales - this is the unintuitive essence of exponential growth and contraction).

I don't consider 10% depletion rate to be small, and I am bemused that you want to start lecturing about exponential curves considering you are the one who tried to suggest that this years north sea oil output decline percentage will naturally exceed last years rate.

Now then, the IEA 2008 report. It doesn't match your numbers. They look at their database of many fields and come up with different natural depletion rates for fields of different sizes, and quite a difference between offshore and onshore. This leads to numbers and timecales that are quite different for their reference scenario than the numbers you are throwing around.

For example here is page 255:

Our oil-production projections are based on a rigorous analysis of trends in output at existing fields (defined as those that were producing in 2007), which is also applied to known fields yet to be developed and fields yet to be found. Among existing fields, some are still building up, some are at plateau and the rest are in decline (see Chapter 10). Thus, the average year-on-year fall in the aggregate production of these fields will tend to accelerate over time, as more and more of them enter their decline phase. In the Reference Scenario, we assume that the decline rate (year-on- year) for all oilfields, once they have passed their peak, is constant for given types and sizes of fields in each region. However, the expected shift in the sources of crude oil, in terms of region, location and field size, means that the average production- weighted observed post-peak decline rate tends to rise, from 6.7% at the start of the projection period to 8.6% by the end.1

In total, output of crude oil from existing oilfields (not including non-conventional projects) drops from 70 mb/d in 2007 to 51 mb/d by 2015 and 27 mb/d by 2030 — a fall of 43 mb/d (excluding projected production using EOR). As a result, 64 mb/d of new capacity, in gross terms, needs to be brought on stream between 2007 and 2030 in order to maintain capacity at the 2007 level and meet the projected 21 mb/d increase in demand. The gross new capacity required by 2015 is about 30 mb/d. Oil production from existing oil sands and extra-heavy oil projects typically declines much less rapidly, as output is largely determined by the accessibility of the deposits and the intensity of mining or steam injection (see Chapter 9).

Production from existing offshore fields declines much more rapidly than that from onshore fields, for the reasons outlined in Chapter 10. On average, output from existing onshore fields falls at a year-on-year rate of 3.2%, from 46 mb/d in 2007 to 22 mb/d in 2030. Output drops much more quickly at existing offshore fields, by an average of 6.3% per year, from 24 mb/d to 5 mb/d. The overall average annual fall in output at existing fields is proportionately much smaller in OPEC countries, at 3.3%, than in non-OPEC countries, where it is 4.7%, reflecting the fact that most OPEC fields are onshore. Production at existing fields falls by 17 mb/d in OPEC countries over 2007-2030, compared with 26 mb/d in non-OPEC countries (Figure 11.3).

The biggest fall in crude oil output from existing fields, in absolute terms, occurs in the Middle East, where some 11 mb/d of capacity needs to be replaced (Figure 11.4). This simply reflects the fact that this region is, by far, the biggest single producing region: the rate of the overall fall in production — an average of around 3% per year — is actually lower than that in any other region. In OECD Europe and OECD Pacific, offshore fields account for most of the loss of oil output; in Asia and Africa, the losses are evenly spread between offshore and onshore fields. Russian and other eastern European/Eurasian countries account for 17% of the overall 43 mb/d fall in output.

 

[Falcon]

I don't consider 10% depletion rate to be small, and I am bemused that you want to start lecturing

I'm not lecturing. A few people read these threads and we talk to a wider audience than just each other. The absurd numbers that small exponents generate is an interesting fact that not everyone anticipates. Relax.

Now then, the IEA 2008 report. It doesn't match your numbers. They look at their database of many fields and come up with different natural depletion rates for fields of different sizes, and quite a difference between offshore and onshore. This leads to numbers and timecales that are quite different for their reference scenario than the numbers you are throwing around.

Yes, which they adjust for.

the worldwide average natural decline rate (year-on-year) rose from 8.7% in 2003 to 9.7% in 2007 (the average rate is 9% for the period 2003-2007 as a whole)
p.245

There is little reason to suppose that, for a given type and size of field in a specific location, the natural decline rate will change significantly in the future
p.247

At the world level, the increase in the production-weighted average decline rate over the projection period is about 1.5 percentage points, taking the rate to around 10.5% per year in 2030
p.248

I'm using 10%.

 

[Falcon]

When you look at the capital directed towards vanity projects such as the Apollo programme, and the continuing capital directed towards the military, you can see the kinds of efforts that can be made where there is the will to do it.

The will to do it. And sufficient surplus energy, relative to the amount required to sustain the system wide functionality upon which massive projects depend, to implement massive projects.

This problem is energy deficit, and therefore impairment of system wide functionality. Surplus energy enables technology. It does not follow that technology enables surplus energy. This is single-point-of-failure territory.

 

[elbows]

I'm using 10%.

Is it at all safe to do that when the detail in the report you praise mentions the middle east declining at a substantially lower rate, and with their production numbers for existing oilfield output to 2030 ending up significantly different to yours?

Im just not sure what the point of stretching these numbers is, given that their numbers are already quite dramatic enough.

Now I am hungry for more recent data, know of any follow-up studies that give us data on how the existing fields they mention have actually declined from 2007-2011?

 

[elbows]

By the way I don't presume that the estimations for slower natural decline rates on-shore in the middle-east are a real safe bet either. But I won't take this idea any further without actual signs of a worse scenario unfolding.

 

[Falcon]

Is it at all safe to do that when the detail in the report you praise mentions the middle east declining at a substantially lower rate, and with their production numbers for existing oilfield output to 2030 ending up significantly different to yours?

The North Sea and Russia are both up around 14% (UK was a 22% fall last year WITH investment). 10% is the production weighted outcome. It is safe enough for the first approximation sufficient to show we haven't a cat's chance of substituting such voracious depletion.

 

[elbows]

The North Sea and Russia are both up around 14% (UK was a 22% fall last year WITH investment). 10% is the production weighted outcome. It is safe enough for the first approximation sufficient to show we haven't a cat's chance of substituting such voracious depletion.

Safe enough for you, not for me. Don't get me wrong, it leans in that direction, but the detail matters. And as you know I would be much happier if, considering you think we have no chance, you'd be prepared to pick a date that you think substitution efforts will fall below the required level to maintain a plateau rather than drop in production.

 

[Falcon]

Safe enough for you, not for me. Don't get me wrong, it leans in that direction, but the detail matters. And as you know I would be much happier if, considering you think we have no chance, you'd be prepared to pick a date that you think substitution efforts will fall below the required level to maintain a plateau rather than drop in production.

Well without resorting to weasel words, they already have. Plateau is already only being maintained by liquidising food we need to eat, funded by money we are printing, and dumping the emissions into an atmosphere that can't accommodate them. If we gave a shit about that half of the world's population which the Food and Agriculture Agency already estimates lives in low-income, food-deficit countries that are unable to produce or afford to import enough food to feed their people, and our kids, we'd be off plateau now.

So a better question is when system functionality will become so degraded that we can't maintain plateau, given our complete disregard for our own welfare. I can't estimate that.

 

[love detective]

can you state in a few simple sentences what the empirical evidence is which shows the causal impacts of oil prices rises on the financial crisis? all you've done so far is to point to a tripling (although your graph shows a doubling) of commodity prices in a three year period. I've taken the numbers that this boils down to (provided by free spirit) and shown that these are a fraction of the impact caused by the unfolding of the subprime lending debacle. Please show me, clearly and concisely why these small impacts had a much large impact than things which had a much larger impact on household finances. You simply haven't done this yet, you've ignored my rebutals of your point and continued to assert your point. Can you do that now for me?

Make your case and point me towards the data which shows that these price inflation increases were the single biggest thing contributing to sub prime borrowers inability to keep up with their mortgage payments? If it's so obvious that this was the cause, the facts & figures must be readily available for you to provide to me now - so please, if you want me to take your theory seriously, back it up

But answer came there none......

 

[Falcon]

But answer came there none......

Read it yourself. Then tell me what you disagree with.

 

[love detective]

so you are unable to do what i asked - state the reasons, in your own words, as to why your so far asserted opinion should be accepted as credible by me

I'll ask again:-

Please show me, clearly and concisely why these small impacts had a much large impact than things which had a much larger impact on household finances. You simply haven't done this yet, you've ignored my rebutals of your point and continued to assert your point. Can you do that now for me?

Make your case and point me towards the data which shows that these price inflation increases were the single biggest thing contributing to sub prime borrowers inability to keep up with their mortgage payments? If it's so obvious that this was the cause, the facts & figures must be readily available for you to provide to me now - so please, if you want me to take your theory seriously, back it up

 

[love detective]

actually just had a flick through that and all but one section of it either focuses on the causes of oil prices rises/sjocks or the consequences of historical oil price shocks - only one section (section 5) focuses on the consequences of the current oil price shock in 2007/2008 - and that seems to mainly focus on the impact on the automobile industry (and even then it doesn't even attribute oil price rises as the key driver for this e.g. 'Although gasoline prices were likely a key factor behind plunging sales for U.S. automakers in the first half of 2008, falling income appears to be the biggest factor driving sales back down in the fourth quarter of 2008')- there's nothing in there pointing towards what you claim the oil price shock did (at least not that i can find through a brief flick through)

Again, if it's so obvious instead of just posting a link to a 70 page report - just quote me a few paragraphs of it where it backs up your point?

can't be that hard can it?

I can pick out a few quotes from it that help my case more that it helps yours, e.g.

Blanchard and Galí are doubtless correct that there has been some decrease in the effects of oil prices as the economy has become less manufacturing-based and more flexible, and since the housing downturn surely made a critical contribution to the recession of 2007-08

I would note first that housing had been exerting a significant drag on the economy before the oil shock

Eventually, the declines in income and house prices set mortgage delinquency rates beyond a threshold at which the overall solvency of the financial system itself came to be questioned

He also makes somes leaps, despite what he says above, that oil is repsonible for everything pretty much doing what you do without actually making the case as to why - he just says 'in my mind' it was oil - not particularly convincing account

 

[elbows]

Well without resorting to weasel words, they already have. Plateau is already only being maintained by liquidising food we need to eat, funded by money we are printing, and dumping the emissions into an atmosphere that can't accommodate them. If we gave a shit about that half of the world's population which the Food and Agriculture Agency already estimates lives in low-income, food-deficit countries that are unable to produce or afford to import enough food to feed their people, and our kids, we'd be off plateau now.

So a better question is when system functionality will become so degraded that we can't maintain plateau, given our complete disregard for our own welfare. I can't estimate that.

Whoa there, oh here we go again! Why are you trying to suggest that right now the unconventional oil equivalents are responsible for maintaining the plateau up till this point? Surely the available data suggests that conventional oil is plateauing up till at least this point, and the unconventional stuff is whats produced the additional 'growth' in energy liquids supply?

Are we talking about two different plateaus or something? Are you talking about the total of all liquids being in plateau, and Im talking about a lower, oil-only figure that is plateauing?

Either way we are back to some territory where I won't disagree with you about many of the conclusions, all the options are disgusting when we consider the plight of the less-well off, but their plight was also a disgrace during the 'glory years' of growth.

 

[Falcon]

He also makes somes leaps, despite what he says above, that oil is repsonible for everything pretty much doing what you do without actually making the case as to why - he just says 'in my mind' it was oil - not particularly convincing account

On the basis of his "flick through", LD offers a total misrepresentation of the paper, skimming anything from which to construct a straw man.

In fact, it uses the price elasticity of petroleum and the value share of energy purchases relative to total expenditures (including his mortgage payments) to assess econometrically the relationship between oil price and economic recession, finding strong correlation (in the statistically meaningful sense) between economic recession and oil supply/demand divergence. A particularly interesting element is his analysis of the role of speculation in oil price (a far stronger basis for rebuttal by LD of my assertion, if he had understood it). He finds there to be some evidence, but notes that the two key ingredients to make such a story coherent - a low price elasticity of demand, and the failure of physical production to increase — are the same key elements of a fundamentals-based explanation of the same phenomenon.

LD is quoting selectively. The full "housing was exerting significant drag" quote is:

housing had been exerting a significant drag on the economy before the oil shock, despite which economic growth continued. Residential fixed investment subtracted an average of 0.94% from the average annual GDP growth rate over 2006:Q4-07:Q3, when the economy was not in a recession, but subtracted only 0.89% over 2007:Q4-2008:Q3, when the recession began. At a minimum it is clear that something other than housing deteriorated to turn slow growth into a recession.

Those factors were consumer sentiment and purchasing power hammered by rising energy prices, and unequal reductions in spending, specifically motor vehicle purchase. The 25% annual fall in SUV sales in 2008, due to falling income rather than rising petrol prices, meant a further reduction in income for those employed in manufacturing and selling cars, amplifying the economic impact of the oil shock on the economy. Moreover, elasticity shifted as the price got higher - the reduction of purchasing and sentiment was much greater in the $100-140 increase than in the $60-100 increase.

Furthermore, had LD's theory been correct (it woz the monthly mortgage premium increase wot dunnit), house prices would have fallen uniformly. Instead:

house prices in 2007 were likely to rise slightly in the zip codes closest to the central urban areas but fall significantly in zip codes with longer average commuting distances. Foreclosure rates also rose with distance from the center. And certainly to the extent that the oil shock made a direct contribution to lower income and higher unemployment, that would also depress housing demand.

Finally:

some degree of significant oil price appreciation during 2007-08 was an inevitable consequence of booming demand and stagnant production. It is worth emphasising that this is fundamentally a long-run problem, which has been resolved rather spectacularly for the time being by a collapse in the world economy. However,the economic collapse wil hopefully prove to be a short-run cure for the problem of excess energy demand. If growth in the newly industrialised countries resumes at its former pace, it would not be too many more years before we find ourself back in the kind of calculus that was the driving factor behind the problem in the first place. Policy-makers would be wise to focus on real options for addressing those long-run challenges, rather than blame what happened last year entirely on a market aberration.

 

[love detective]

Falcon you should go and do a speaking tour of the US and tell all those millions who lost their homes that it wasn't because they couldn't afford the massively increased mortgage payments that did it - i'm sure your theory will be comforting to the harsh reality that they actually faced due - i'm sure you'd get short thrift trying to explain your theory to them as to why they lost their homes and what were the causes

 

[Falcon]

Whoa there, oh here we go again! Why are you trying to suggest that right now the unconventional oil equivalents are responsible for maintaining the plateau up till this point? Surely the available data suggests that conventional oil is plateauing up till at least this point, and the unconventional stuff is whats produced the additional 'growth' in energy liquids supply?

No. Definitions of conventional and unconventional oil varies a bit between agencies. But taking the figures in EIA International Energy Outlook 2011, total production in 2011 was 88 million barrels, comprising 71 million barrels of conventional and 17 million barrels of unconventional. They forecast conventional oil falling to 48 million barrels by 2035, with the balance of total projected demand of 127 million barrels - a whopping 79 million barrels a day -- being supplied by unconventional oil for which no historical technical precedent exists.

Conventional oil peaked in 2005. I'm not suggesting that unconventional oil equivalents are responsible for maintaining the plateau up till this point. I'm quoting institutional data to that effect.

 

[littlebabyjesus]

Not everyone saw an increase in their mortgage. Ld's point was specifically about subprime. And by its nature, subprime was not in the most desirable areas. House prices rarely fall uniformly. Look at the uk. There has been a similar stretching of the market, prices at the bottom going down and those at the top staying level or going up.

That picture of house price falls being concentrated in less desirable areas is exactly what you would expect to see with the collapse in subprime.

 

[Falcon]

Falcon you should go and do a speaking tour of the US and tell all those millions who lost their homes that it wasn't because they couldn't afford the massively increased mortgage payments that did it - i'm sure your theory will be comforting to the harsh reality that they actually faced due - i'm sure you'd get short thrift trying to explain your theory to them as to why they lost their homes and what were the causes

Fallacy of false dichotomy. It wasn't either/or. It was both. We are discussing here the primary cause of both, to better understand rational responses.

 

[Falcon]

Not everyone saw an increase in their mortgage. Ld's point was specifically about subprime. And by its nature, subprime was not in the most desirable areas. House prices rarely fall uniformly. Look at the uk. There has been a similar stretching of the market, prices at the bottom going down and those at the top staying level or going up.

That picture of house price falls being concentrated in less desirable areas is exactly what you would expect to see with the collapse in subprime.

You are suggesting that the suburbs of US cities are less attractive than the inner cities? Have you travelled much in the US?

 

[littlebabyjesus]

You are suggesting that the suburbs of US cities are less attractive than the inner cities? Have you travelled much in the US?

Generally yes, I would expect so. A family home in a city centre will cost a lot more than one in an outer suburb, generally. That's how it is in the uk. I'd expect it is similar in the us. Far far far cheaper to live in jersey city and commute to manhattan than to live anywhere in manhattan.

 

[Falcon]

Generally yes, I would expect so. A family home in a city centre will cost a lot more than one in an outer suburb, generally. That's how it is in the uk. I'd expect it is similar in the us. Far far far cheaper to live in jersey city and commute to manhattan than to live anywhere in manhattan.

No. In the sorts of places where the largest concentrations of subprime mortgages were issued (the effect you correctly flag), you get shot in the face in the centre of the city. Buy a house in the middle of Detroit.