1. Did a financial crisis cause a fall in expected and actual aggregate demand? (With central banks being unable or unwilling to do enough to stop it).
2. Or did a fall in expected and actual aggregate demand cause (or worsen) a financial crisis? (With central banks being unable or unwilling to do enough to stop it).
Obviously, there is a positive feedback multiplier between financial crises and expected and actual aggregate demand growth. (If central banks are unable or unwilling to prevent there being positive feedback.)
I started to write a post on this, but scrapped it, because I couldn't think of any way to provide good evidence one way or the other.
Now David Beckworth has a very important post which provides evidence for the second hypothesis. Go read it. Or simply look at the picture. US households' expected dollar income growth started to fall well before the US financial crisis.
[Update: see also Marcus Nunes' post and graph.]
The purpose of this post is simply to draw attention to David's post, and to raise a few minor points:
1. It's not enough just to look at the US. The same thing happened in a lot of other countries at the same time.
2. Households' average expectations of their own nominal household income growth isn't exactly the same as expectations of nominal GDP growth.
3. You can see the 1982 and 2009 recessions clearly foretold in David's picture. But you can't see, or can only just see, the 1991 and 2001 recessions.
4. David's graph starts to head south in 2005. It's just a bit too early for comfort, if you want to say that declining expected nominal income is what caused the 2008 financial crisis and 2009 recession.
5. Putting 3 and 4 together: while it is a commonplace in intertemporal macroeconomic theory that a decline in expected future nominal income would cause a decline in current desired nominal expenditure ("aggregate demand"), that can't be the only thing that affects aggregate demand. Interest rates, for example, would be another thing. Can those "other things" fill in the small gaps between David's picture and what actually happened?
6. "Never reason from an expectations change". What caused/allowed expectations to change like that?
In any case, Scott Sumner's crazy idea that declining actual/expected NGDP is what caused (at least a large part of) the financial crisis, rather than vice versa, is looking a lot less crazy. And maybe it's time to put macro into finance, rather than (just) vice versa.
Gotta go grade exams, and do a lot of admin stuff.
Worthwhile Canadian Initiative 
Well, I don’t see possible to separate so easely the real from the Financial. The big problem is acumulated Debts, and I don’t believe that more proactive FED could have reduce the crisis to a simple recesion. It is irrelevant for me if the falling demand was first than financial crisis.
There was also a commodity price shock especially the price of oil. The recessions of the mid and late 70’s were in part driven by the same phenomenon. The housing bubble burst around 2006 and I believe was contributing negatively to gdp thereafter.
(Not an economist.) There was a big run-up in oil prices in 2007-2008, causing a noticeable increase in transportation costs in North America. Could that, coupled with the trends in household expectations, have triggered the 2008 crisis? Or is this too specific to factor into the model you’re describing?
Nick, I updated the post with figures showing NGDP and recessions. The paper I link to in the post looks at expected income and changes in wealth to explain the downturns. It finds these two alone explain most of the decline. Here is the link to the article: [Link here NR]
Dean Baker has commented on this subject before. As usual it is hard to find anything to disagree with in his opinion…..
Click to access origins-of-the-financial-crisis-5-4doc.pdf
The only weakness I can find in Dean’s arguments is that the housing bubble itself might have masked an employment crisis that would have been inevitable with such a mismatched balance of trade. Spain looks to have been the same.
Yes, Spain looks the same. The same bubble and similar bust. The same excess, imposible to dominante with monetary Policy. But the Big diference of course is the big mistrales of ECB after the crisis.
I suppose that the Sumner’s “crazy” Theory requires that there was no bubble at all previus to the crisis. If We asume y bubble and its consequences in huge Debt, we can explain the magnitude of the crisis. If not, We most suppose that all the fault is for the FED.
What did happen in 2008 was that the bubble finally did burst. That was predictable and predicted.
There is a book from Warren Brussee: “The Second Great Depression, Starting 2007, Endng 2020” printed in 2005, based on data until 2003. The guy is still alive, and even has a blog.
That was before the subprime and before the house bubble in the US really took of.
The seeds for that were sawn in the 1990ties, 1998 the latest, by a Greenspan running interest rates too low.
To start his plots at 2005 means that David Beckworth still doesn’t have a clue.
I have read Professor Sumner’s arguement and discussed it at his blog site.
Prof’ Sumner defines money as $100 bills (that is not an exageration). The professor would say that the market crashed because of rational expectations of an ecnomic slow down due to insufficient base money cration by the fed.
I would include most financial assets as money. The rapid fall in asset prices was a destruction of money. The economy slowed because of the crash. The market crashed because markets are unstable. They are not irrational, but they are subject to feedback loops which make them chaotic.
There is also insufficient discussion of what falling asset prices do to debt burdens. Asset prices fall debts stay the same. One of the cures of the malaise is to reduce the real debts. The ways to reduce debts are to pay it off, inflate it away, and default. There has been so much policy effort to prevent bankruptcy and default. We should be encouraging it!
“…while it is a commonplace in intertemporal macroeconomic theory that a decline in expected future nominal income would cause a decline in current desired nominal expenditure (“aggregate demand”)…”
I’m trying to get my head around the story David Beckworth is telling re the sources of the recession. Circa 2008, a typical respondent to the Michigan survey was thinking what? “I need to curtail my present expenditure because I expect the Fed will soon adopt a new policy stance that will reduce the rate of future NGDP growth. The rate of price-inflation is not going to fully adjust to reflect the Fed’s new stance immediately, so I expect my real income to be below-normal over the net few years. I can’t borrow against my future income, so if I want to maintain my customary level of real consumption over the transition period I need to build up a little nest-egg for this purpose starting today.”
Is this even close?
“Households’ average expectations of their own nominal household income growth isn’t exactly the same as expectations of nominal GDP growth.”
First, I think you need to get the term “wages” in there somewhere.
Second, falling wage share of national income. See my comment here.
http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/04/begging-the-long-run-question-inflation-variability-vs-ngdp-growth-variability.html?cid=6a00d83451688169e201901b84beb5970b#comment-6a00d83451688169e201901b84beb5970b
You need to do monthly budgeting. For most of the lower and middle class, prices were rising more than wages, and they were using currency denominated debt to make up the difference. Currency denominated debt was being used to prevent price deflation. Price inflation measures overemphasized tradable goods. China was exporting price deflation in tradable goods, but not in college tuition, health insurance, energy, etc.
Brussee made his prediction based on a comparison to Japanese Consumer data.
Japan built their financial time bomb in the 60ties and 70ties, with interest rates pressured low, and then it exploded into their face 1990.
That looked in 1985 like a clever move to give the Japanese industry an advantage compared to the US and Germany (Queisser: Kristallene Krisen 1985. I loved his black sheep tie at the Punktdefekttreffen : – ).
But what they had to do after 1990, kind of following the Krugman advice, has brought them now into such a desparate situation, that they made now their financial “Pearl Harbor decision”
http://www.ft.com/intl/cms/s/0/894422b4-a5d6-11e2-9b77-00144feabdc0.html#axzz2RPEqIIom
We learned from Izzabella Kaminska over Xmas, that it took the Roman Empire 60 years after debasing the currency to then let the inflation destroy the core of their economy.
“Never reason from an expectations change”. What caused/allowed expectations to change like that?”
Reality! People saw their jobs being exported to China, saw their jobs being taken by machines, and saw things like if you go on strike, firms will replace you with someone else. They should have seen this back in the 1990’s.
“In any case, Scott Sumner’s crazy idea that declining actual/expected NGDP is what caused (at least a large part of) the financial crisis, rather than vice versa, is looking a lot less crazy. And maybe it’s time to put macro into finance, rather than (just) vice versa.”
About the finance part, can banks and bank-like entities increase purchasing power on goods/services and/or financial assets? Yes/No
Luis said: “The big problem is acumulated Debts”
I agree, but good luck getting Scott, Nick, or most other economists to say too much currency denominated debt (both private and gov’t) is the problem.
You will need to start by defeating the argument that for every borrower there is a lender.
Yes, but this argument IS true only accontably. It doesn’t mean anytning if IS so dificult to reestructure debt.
Sorry, but interest rates were to high back in the 90’s and 80’s which caused leverage to rocket. Then everybody wanted a piece of it.
Debased currency my rear end. Classic backwards look.
When interest rates are to high leverage and debt surge in the system. It wasn’t to mid-2002 when central bank policy finally got loose enough that leverage deceleration started. The leverage boom of 99-2002 was when it was the worst.
David Beckworth claims the basic problem is ordinary households’ inability to deal with debt, which in turn has cut household spending. And his solution? It’s to have the Fed print money and buy up assets owned by the rich. Why the alleged solution has any big effect on the problem eludes me.
genauer: “To start his plots at 2005 means that David Beckworth still doesn’t have a clue.”
He started his plots in 1978. Which is obvious, if you look at his picture.
Giovanni: current desired consumption depends on current income, but also on expected future income. That’s one mechanism. Current desired investment depends on expected future demand, and future demand will (roughly) equal future income. That’s a second mechanism. A third might be that willingness and ability to borrow would depend on expected future income.
Ralph: debt is never a problem. Debt/income ratio can sometimes be a problem. Since most debt is nominal (unindexed to inflation), at the aggregate level that means debt/NGDP ratio can be a problem. There are two ways to reduce a ratio: reduce the numerator; increase the denominator. Since we want to increase the denominator anyway, to get NGDP back up to trend, to reduce unemployment, that seems the sensible way to go.
Nick, Thanks for your response. If more QE really does increase real GDP and hence bring unemployment down, I’m all for that. But I’m puzzled as to why giving the wealthy cash in exchange for government debt will greatly increase their spending: they make a small capital gain as a result of QE, but against that, their marginal propensity to consume is low. Plus the REAL PROBLEM (if David Beckworth is right) is lack of confidence amongst LESS WELL OFF households. QE gives that lot no inducement to spend.
But wages/(household debt) surged between 2000-2006. (Debt service also increased strongly and was not strongly correlated to short-term interest rates (i.e. mostly mortgage debt).) One could argue that households were beginning to experience or had expectations of their own financial crisis, which played into the GFC later. With that debt surge, I think it’s still hard to argue the direction of causality. Each homeowner making the decision to “cut-back a little” because they just bought a big house with new appliances, makes sense locally, but not in aggregate.
“…current desired consumption depends on current income, but also on expected future income. That’s one mechanism. Current desired investment depends on expected future demand, and future demand will (roughly) equal future income. That’s a second mechanism. A third might be that willingness and ability to borrow would depend on expected future income.”
Think of a classical macro world, with across-the-board instantaneous price adjustment and national income perpetually at potential. In such a world why would suddenly realizing the CB intended to reduce the rate of NGDP growth – which is to say, the rate of inflation – in the near future lead me to curtail my current expenditure? So future nominals will be less than I’d previously expected – la-di-da. As long as future relative prices and my future real income are unchanged no need to revise any of my plans. (Yes, there are all kinds of mechanisms by which reducing the expected inflation rate might elicit a current expenditure response – for example, by reducing the effective rate of taxation on investment income or alleviating the front-end-loading of standard loans – but these effects are likely to be (1) largely expansionary and (2) much smaller in size than what we’re talking about here.)
So, if we can’t get much expenditure action in a classical macro world, we must be talking about a world with enough price/wage rigidity so the CB expected actions will engineer a recession and thereby reduce my future real income…the anticipation of which leads me to curtail my current expenditure.
BTW Isn’t a relative reduction in wages exactly what Sumner has prescribed, to cure the sticky wage problem? Shouldn’t we expect nominal wages to lag NGDP to boost employment? In fact David’s chart shows NGDP growth exceeding expected income growth 2000-2008.
Nick,
I would recommend a link between your ideas here and Gary Gorton’s work on the crisis. He has argued that the crisis was an old fashioned bank run but, most interestingly, he does not believe that Lehman’s crash and the ensuing crisis had much of anything to do with the “subprime lending mortgage housing crisis.” In other words, he is halfway to your own argument! I believe that the smoking gun in favor of your position would be to show that Lehman’s crash and the general financial crash were AT THAT MOMENT caused by falling NGDP expectations. It would be interesting to go through Gorton’s work and see if a statistical connection can be detected between his description of Lehman’s crash and tight money from falling NGDP expectations. In other words, can it actually be shown statistically that Lehman’s crash was caused by the phenomenon that Beckworth describes in his post?
Here’s a post by Arnold Kling about this…. http://econlog.econlib.org/archives/2012/07/post_2.html
Best regards.
Nick,
You write:”…current desired consumption depends on current income, but also on expected future income. That’s one mechanism. Current desired investment depends on expected future demand, and future demand will (roughly) equal future income. That’s a second mechanism. A third might be that willingness and ability to borrow would depend on expected future income.”
Alright, but…think of a classical macro world with across-the board instantaneous price adjustment and national income perpetually at potential. Suppose I suddenly realize the CB intends to reduce the growth of NGDP – which is to say, the rate of inflation – in the near future. La-di-da…since this will have no effect on future relative prices or my future real income I have no reason to reduce my current expenditure – or change any of my plans, for that matter. (Yes, there are all kinds of mechanisms – e.g., reducing the effective rate of tax on investment income, alleviating the front-end-loading of standard loans – by which a lessening of expected inflation might itself affect current expenditure. But these effects are (1) largely expansionary and (2) really small compared to the expenditure change we’re talking about here.)
So, if we can’t get much current expenditure action in a classical macro world, we must be in a world with price/wage rigidity – enough rigidity in fact so that the CB’s expected cutailment of NGDP growth is also expected to induce a major recession, thereby reducing my expected future real income. It must be my anticipation of these future income losses that induces me to curtail my current consumption/investment expenditure, thereby shifting the IS curve to the left and moving the recession forward to the here-and-now.
What else could the story be?
From 1992-2000 real median income increased. From 2000- 2004,it fell. It wasn’t irrational to expect that this reverse was temporary and to keep up consumer spending by borrowing. Borrowing increases NGDP by transforming credit, which doesn’t appear in NGDP into investment that does. Since GDP = GDI = GDE, all sides of the equations increase in lockstep.
For every borrower, there is a lender, but the loan is an asset to the lender. There’s no reason for the lender to reduce spending given sufficient liquidity. Increased credit has to increase NGDP. If it didn’t then there would be no point in borrowing to invest, since there could be on average no return on investment. This view of credit captures the fundamental distinction between solvency and liquidity, which a pure loanable funds model doesn’t.
If additional credit increases NGDP. the net repayment (or default) should correspondingly decrease it. Since 2007 consumer credit has decreased even more than government debt has increased, so that total US debt has actually decreased from 2007-2012.
The US is one of the only developed countries that has actually practiced total economy austerity. The European countries have decreased government spending, and had to borrow to cover the revenue shortfall while achieving no decrease in private sector debt. Their total debt has increased. This is an odd sort of austerity – all pain and no gain.
I think, among other things, that this shows the danger of using NGDP as a representation of the entire economy. It isn’t designed for this.
You can also look at total debt / NGDP as a crude measure of financial asset leverage over the real asset economy. We could probably come up with better ones if economists felt it was important enough.
BTW because of the velocity of money, the credit affecting GDP should be 1/2 the increase in the current period + 1/2 the increase in the previous period.
Nick,
I seem to have a comment lost in space. It’s about the trend in median income 1992 – 2000 – 2005 related to NGDP and debt. I’d rather not recreate, if I don’t have to.
Peter: Stephen has retrieved it. And I just retrieved a couple of Giovanni’s.
I’m a bit too tired to respond to comments. But I’m reading them.
Sorry, but there’s probably some redundancy in mine…I tried once and then again when the first didn’t show up.
Among other things this happened in 2008
M4- excludes T-bills. The M4, M4- divergence looks like a flight to safer assets.
I thought this was common knowledge — I.e. the bursting of the housing ponzi scheme clearly implies #2). The crisis was a banking crisis, which was the result of the deterioration of the bank’s assets, namely mortgage loans. This happened as a result of the declining house prices from their bubble values and the corresponding decrease in residential investment, as no one is going to pour money into an asset class whose overall value is going to decline. It was not the other way around — e.g. a banking crisis happens for no reason whatsoever, causing house prices and consequently residential investment to decline.
In terms of what the CB could do to prevent house prices from rationalizing, they could cut rates to zero right away and lift leverage requirements and capital adequacy requirements. That might have been able to distort house prices even further and delay bank recognition of losses for a bit.
For #4 try this:
Tight credit
derivative market crash
durable goods orders show some combination the expectations of producers of consumer goods and availability of credit to them. In either case they change their orders for means of production indicating either they expect less consumer demand or they can’t finance their production or both.
“Since we want to increase the denominator anyway, to get NGDP back up to trend, to reduce unemployment, that seems the sensible way to go.”
There should be a model where raising prices causes real GDP to fall lowering employment. If real wages being too low (probably negative) caused the recession then trying to make real wages negative by raising prices may make the situation worse. I also don’t believe trying to make firms more profitable thru price increases will help employment and/or wages when corporate profits as % of GDP are at or near a record high.
About real earnings being negative:
http://advisorperspectives.com/dshort/updates/Median-Household-Income-Update.php
I believe price inflation measured by a lower/middle person’s budget is about 1% to 2% higher than CPI. That would make the charts look worse. I think that is where Peter N got his first chart.
“What happened in 2008?”
It was a currency denominated debt crisis that led to the amount of medium of account (MOA)/medium of exchange (MOE) to fall creating a shortage.
Ralph Musgrave said: “But I’m puzzled as to why giving the wealthy cash in exchange for government debt will greatly increase their spending: they make a small capital gain as a result of QE, but against that, their marginal propensity to consume is low.”
Exactly. The central bank buys some bonds from a bank. The bank buys some bonds from Apple and Warren Buffett. I don’t see either one spending to consume or spending to invest, just some financial assets being exchanged (the demand deposits just sit in Apple’s and Buffett’s checking accounts with zero velocity in the real economy). If lower interest rates (the central bank will probably overpay) don’t get the savers to spend and no one else goes into currency denominated debt, then the economy is basically going nowhere. In other words, both Apple and Buffett want to continue to save (not consume and not invest).
“Plus the REAL PROBLEM (if David Beckworth is right) is lack of confidence amongst LESS WELL OFF households. QE gives that lot no inducement to spend.”
It is not about confidence. It is about reality smashing their previous incorrect assumptions and straining the monthly budget.
Leo said: “There was a big run-up in oil prices in 2007-2008, causing a noticeable increase in transportation costs in North America.”
I’d say it contributed. That was one price going up more than wages, straining monthly budgets. It also affected the budgets of other entities.
home equity, consumer debt and federal debt. Fall in equity precedes fall in debt. Increasing federal debt effectively transfers the debt to the public sector. Total debt decreases. This is austerity that at least sort of works. In Europe you have increasing government debt and increasing consumer debt. That’s austerity that doesn’t work.
Peter N, why not TCMDO (Total Credit Market Debt Owed) at the St. Louis FRED?
Click to access z1r-2.pdf
http://research.stlouisfed.org/fred2/data/TCMDO.txt
What happened in 2008?
A build up in total debt that burst at the seams –
http://research.stlouisfed.org/fred2/graph/fredgraph.pdf?&chart_type=line&graph_id=&category_id=&recession_bars=On&width=630&height=378&bgcolor=%23b3cde7&graph_bgcolor=%23ffffff&txtcolor=%23000000&ts=8&preserve_ratio=true&fo=ve&id=TCMDO_GDP&transformation=lin_lin&scale=Left&range=Custom&cosd=1960-01-01&coed=2013-01-01&line_color=%230000ff&link_values=&mark_type=NONE&mw=4&line_style=Solid&lw=1&vintage_date=2013-04-29_2013-04-29&revision_date=2013-04-29_2013-04-29&mma=0&nd=_&ost=&oet=&fml=b%2Fa&fq=Quarterly&fam=avg&fgst=lin
Your wish is my command.
total debt and its components. Note that it is going down. This is true for very few developed countries (Canada, Korea and Australia IIR).
You can imbed with and img tag src=”your url here”. If you have a FRED account, you can save graphs and get a link for them.
As part of what happened in 2008 note in the graph above the huge drop in financial sector credit. That has to be extremely contractionary. And it’s still going down. It’s amazing the economy is doing as well as it is.
I will give it a shot:
and img tag src=”http://research.stlouisfed.org/fred2/graph/fredgraph.pdf?&chart_type=line&graph_id=&category_id=&recession_bars=On&width=630&height=378&bgcolor=%23b3cde7&graph_bgcolor=%23ffffff&txtcolor=%23000000&ts=8&preserve_ratio=true&fo=ve&id=TCMDO_GDP&transformation=lin_lin&scale=Left&range=Custom&cosd=1960-01-01&coed=2013-01-01&line_color=%230000ff&link_values=&mark_type=NONE&mw=4&line_style=Solid&lw=1&vintage_date=2013-04-29_2013-04-29&revision_date=2013-04-29_2013-04-29&mma=0&nd=_&ost=&oet=&fml=b%2Fa&fq=Quarterly&fam=avg&fgst=lin”
It’s an html tag just remove the blank between “<” and “img” in
< img src=”http://…..”>
Trying now:
Try something shorter. It not only has to be valid HTML, but Typepad has to like it, and I have no documentation of the feature. I just saw someone use it and looked at their HTML source to see what they had done. I have no idea what the restrictions are on things like URL length. Ask FRED to generate a link. It’ll be maybe 10% as long – like
http://research.stlouisfed.org/fredgraph.png?g=hVE
Try a preview with my URL. If it works for mine and not for yours, then you’ll know where the problem lies.
I think there may be some value in having these charts together like this. This one is from one from FT Alphaville that DeLong included in a paper he put up today on 2008 and fiscal policy. The subject of the chart is safe assets.
Peter N said: “For every borrower, there is a lender, but the loan is an asset to the lender.”
How are you defining borrower and lender?