Productivity & Effective Demand: An Intriguing and Disturbing Story . . .
Edward Lambert at Effective Demand; Effective Demand = Effective Labor Income/(cu*(1-u)) points to the result of an economy left to maximize Profits at the expense of Labor. I have my own version or underlying causes of this issue and Edward gives the economic side of it.
I am going to show a graph of Productivity against Effective demand. It is an intriguing and a disturbing graph. Let me start by giving the equation for the productivity used in the graph.
Productivity = real compensation per hour: business sector/(labor share: business sector * 0.78)
The data for this equation comes from this graph at FRED.
The equation for effective demand is…
Effective Demand = real GDP * (labor share: business sector * 0.78)/TFUR
TFUR (total factor utilization rate) = capacity utilization * (1 – unemployment rate)
Let me just show the graph and then start explaining . . .
The graph shows quarterly data from 1967 to the 1st quarter of 2013. The red dashed line is a trend line for the data. We can see that from 1967 to 1997, the plot stayed very tight on the trend line. There were deviations from this line during times of shocks and recessions. But it is very interesting how closely the plot followed the trend line before 1997.
Before 1997, the plot going below the trend line was associated with a recession. The explanation of this is that effective demand rises more during a recession because of more available capacity of labor and capital. At the same time. productivity tends to fall behind the trend line due to rising labor share, not falling real compensation.
When the plot goes above the trend line, productivity is ahead of effective demand. Productivity rises due to labor share settling down and real compensation rising. Effective demand tends to stay still during the expansionary phase of a business cycle. The economy grows up to the effective demand limit and then gets set for a contraction.
We used to have a balance between productivity and effective demand. The economy moved directly on top of the trend line for many many quarters. And now the economy has lost that balance. Since the late 90’s it is a fleeting moment when productivity and effective demand come together on the trend line.
Before 1997, there was very little movement away from the trend line. Then something unusual happened between 1997 and 2001, the dotcom bubble years. The plot went progressively below the trend line even though there was no recession. Productivity was rising during these years, but effective demand was rising at such an unusual rate that productivity could not keep up with it. Effective demand was being artificially created and inflated. The recession of 2001 followed the same unusual path as before the recession.
In 2002, the economy had to make an adjustment. Productivity had to rise or effective demand had to fall. During the housing bubble years (from 2002 to the quarter right before the 2008 recession), productivity rose, while effective demand basically stayed steady. The plot went back above the trend line showing that productivity was beyond the capacity of effective demand and that productivity was at a non-sustainable level. The economy sustained this high level of productivity in the face of low effective demand for a few years, but eventually the correction would come in 2008. The correction was a collapse.
Look at where the economy is now. Since the end of 2010, the plot has barely moved from a productivity just below 1.4 and an effective demand around $14.1 trillion. The plot is way above the trend line and has been just sitting in the same spot for over 2 years. Effective demand is too low for the current productivity in the US. This is an economic bomb building energy that will eventually go off when real GDP approaches $14.1 trillion.
A friend of mine had a dream a few nights ago, where a spirit said that the economy is dying. The graph above would lead one to think the same.
Think about it… where can the economy go now from here?
There are 2 options . . .
Option #1 looks at the equation for Productivity: You have to lower productivity by increasing labor share in relation to real compensation.
1. If you lower real compensation, labor share would fall, but it would have to fall slower than real compensation. Keep in mind though that a lower labor share would lower effective demand too, which would work against the objective. However, if labor share actually rose in the face of lowering real compensation, you would see an economic contraction. So lowering real compensation is not a good option.
2. On the other hand, if you raised labor share faster than raising real compensation, productivity would come down as effective demand increased from higher labor share. This is a safe and sensible way to correct the huge imbalance we find ourselves in.
Option #2 looks at the equation for Effective Demand: You have to increase effective demand back up to $16 trillion. There are two options here as well.
1. Utilization of labor and capital would have fall. (TFUR in the equation above would have to fall.) This would mean a rise in unemployment, which would mean another collapse.
2. Labor share would have to rise. This would also have the beneficial effect of lowering productivity, as long as real compensation rose moderately.
As we can see, the only real option to avert another collapse is to raise labor share of income. This is not likely as businesses are even now fighting an increase in just the minimum wage. Businesses are trying to maximize their profits and do not want to raise labor costs. Yet this objective of theirs is going to kill the economy.
The graph above shows that there is a bomb ticking, and it is a bigger bomb than we saw in 2008. Higher productivity in the face of low effective demand is unsustainable. Yet, we have been sustaining it for over 2 years now with an incredible expansionary monetary policy in the face of an incredibly low labor share of income.
In other words, the economy does better when the workers earn more money. What a “novel” concept!
Jerry:
Partially right; but, there is more than just your observation.
Jerry
thanks. i was about to write a long comment explaining why i couldn’t understand it.
What is the colon operator?
kalesberg:
A litle more direct please. Checked for “misspleled” words already.
addendum: Sorry Kalesberg; I did not have my mathematical head on to answer you. 🙂
“Before 1997, there was very little movement away from the trend line. Then something unusual happened between 1997 and 2001, the dotcom bubble years.”
We ran surpluses. Each stretch away from the trend line was accompanied by declining deficits as a % of GDP. Same with period leading up to 2008. The trade deficit as a % of GDP is also a factor.
Look at net private saving (% change semi-annual):
http://research.stlouisfed.org/fredgraph.png?g=kfG
run
regarding the colon operator: in your definition of effective demand there is a colon. it is hard to tell what the colon is saying. i was going to try to figure it all out (used to have an i.q. around here somewhere), but there were too many undefined terms for me to even get started.
there is probably some use to quantifying and maybe even identifying some factors that justifies the abstraction and complication of all this, but if so you need to identify your audience and decide how much help they need in knowing what you are talking about. in my case it would be a lot.
Coberly:
I will let Edward Lambert explain it as it is his article and I would not want to misniterpret.
Last chart to add is Federal Government Expenditure growth YoY 1967 to 2013 – hopefully I pulled the correct stat at FRED. Growth was not good over the periods of poor effective demand. Well less government expenditures does effect aggregate demand.
http://research.stlouisfed.org/fredgraph.png?g=kgg
oh, c’mon, run. be bloody bold and resolute. you brought this to us. you must have some idea what it means.
here are my questions in order of occurence:
“I am going to show…” who is “i”? run says it’s lambert, but no indication of a quote is given.
“Productivity = real ….” A formal definition is nice. It would help to have an “informal” definition so we have some idea what the author is talking about. Same with “real compensation” Not sure what the “colon” is telling us, or what “business sector means, or why we are multiplying (?) it by 0.78. and if we are dividing business sector by business secor time 0.78, the answer will be 0.78. What does that tell us?
“Effective demand..” again, hard to tell what this means or why.
“TFUR..” again, a few informal remarks might help us know what the abstraction is getting at. what is “capacity utilization” and why are we multiplying it by what looks like it might be “capacity utilization” that is what we have after subtracting the unemployment rate.
All this might be perfectly clear to those who took the same class or read the same book, but that is an unreasonable assumption in the bigger world.
the graph appears not to be meant to be a graph “over time,” but neither is it a “function.”
what we appear to have is some sort of relation (?) between “productivity” , defined according to an obscure relationship, to “effective demand” also defined somewhat obscurely.
“The explanation of this…” Well, the explanation is not blindingly obvious, even after you explain it.
I had to stop here. I can’t form any mental picture of what we are talking about, and that makes it really hard for me to follow the argument.
Let’s look at the equation for Effective demand…
Effective demand = real GDP * labor share/(labor * capital utilization)
Effective demand simply multiplies real GDP output by the ratio of effective labor share to labor and capital utilization. There are many factors that affect the utilization of labor and capital. How the income is spent or saved? How the govt runs a deficit or not? But those are details outside the equation. However money is spent and by whom, labor and capital adjust to the demand.
When effective demand is far above real GDP, utilization of labor and capital decrease. When effective demand comes back down, it compresses the dynamics of utilizing more labor and capital.
After the crisis of 2008, effective demand sky-rocketed from low utilization of labor and capital. Then it settled back down quickly during 2010, the same time that productivity stalled out. You can see that in the graph above. (note: the thought of the day is that there is a limit line above and below the trend line for productivity and the plot cannot go beyond it. In effect there is an upper limit from demand on productivity. That limit line must have an equation.)
When we go back to the late 90’s, we see the opposite. Effective demand began to inflate and real GDP followed it into bubble territory. Effective demand led real GDP growth. Labor share rose while unemployment was low. Real wages were increasing. Productivity increased but at a slower pace than effective demand. Thus, the plot line scooted below trend line. That scenario had never been seen since 1967 in the graph above.
Why did effective demand increase? Production increased but productivity did not increase as much. On top of that, labor share rose and real wages were rising. Demand was increasing, production was increasing, but productivity was increasing at a slower pace.
When the 2001 recession hit, the plot line was already below trend line and just scooted out more along the same bubble path. Kind of weird actually.
What I find really interesting is the difference between the dotcom bubble years and the housing bubble years. During the dotcom years, effective demand increased a lot while productivity lagged behind. Then during the housing bubble, the already inflated effective demand stayed level, while productivity increased. The result was a correction from below the trend line to above the trend line. Very different dynamics between the two bubbles. One created effective demand, the other consumed it through productivity, stagnant real wages and a declining labor share.
Matt’s comment about govt surpluses and trade deficits is important to talk about. It seems government surpluses during a boom time increased private production. The trade deficit which started to grow around the beginning of the dotcom bubble created an inflow of capital for investment to increase production as effective demand was growing. I would like to hear more of Matt’s views on that.
Coberly,
I hear you. These are new concepts.
The trick then is to understand effective demand, especially when one is trained in thinking about aggregate demand. They are very different.
The best way to think about effective demand is the unused capacity of demand in the economy. During the depth of a recession, effective demand increases very high, which may confuse you. It is only showing the capacity of demand is great from low production, and that businesses will be able to bounce back. At the top of the business cycle, when effective demand comes down to real GDP, unused capacity of demand is low, and businesses will not grow like they did from the bottom of the recession.
Real GDP can still grow with low effective demand, but the dynamics for business are different from when there is higher effective demand.
The colon is just to show the data is Business sector as opposed to farm, nonfarm and the like. It is not a colon operator.
0.78 is used because it is the central tendency line with a y-intercept of 0 between capacity utilization and labor share (business sector, 2005=100). There is a growth model that explains the reason for the y-intercept of zero.
TFUR is simply multiplying the capacity utilization rate by the labor utilization rate. For example, capacity utilization = 75% and unemployment = 8%. The TFUR would be (0.75 * 0.92) = 0.69.
The normal definition of productivity is Y/L (output/total labor hours). OK… Now we look at labor income which is equal to real wages per hour (W) * total labor hours (L). Labor income is also equal to real output (Y) * labor share (e). Thus we have…
W*L = Y*e
We arrange to get productivity…
Y/L = W/e
maybe “W/e” is an obscure definition of productivity, but it is a valid one.
The graph reminds me of this one.
http://angrybearblog.strategydemo.com/wp-content/oldimages/angrybear/1/-mYa96__tEx0/T4XnYhW8COI/AAAAAAAAByg/SiAR3E1QXiA/s1600/Consumption_vs_Net_Worth.JPG
Which was part of this post.
http://angrybearblog.strategydemo.com/2012/04/another-look-at-wealth-and-consumption.html
I like the conclusion here, but am perplexed by the methodology. I might have to reread this several times for it to sink in.
But I am deeply troubled by having output show up in the demand definition, but not in the productivity definition.
This seems totally backwards to me.
Cheers!
JzB
Lambert
Thanks. I hope I am not being too stupid here. You may be saying something interesting and important. But before I can even get to your argument, I need to understand your terms, and I am having a hard time doing that.
It may be that I lack “basic” economics education. It might be that you are so familiar with your terms you don’t realize they need a little unpacking for the lay reader.
I’ll look again when I get some more time.
JzB…
Fascinating graph from that post in 2012. The curves look extremely similar.
I updated the graph here…
http://research.stlouisfed.org/fredgraph.png?g=kgY
Recently net worth is rising at a faster rate than personal consumption. Is this another definition of a bubble with rising asset values, while consumption lags behind?
The effective demand graph has the same shape for the curve but at different times. During the housing bubble, effective demand was constant while net worth grew. Then during the crisis, effective demand grew while net worth fell. But in the end they ended up with similarly shaped curves.
Currently effective demand is NOT increasing, while the graph you bring up shows net worth increasing. In the past they moved together, supposedly in balance. The current difference is troubling for me because it is the same scenario before the crisis of 2008.
Coberly,
You are absolutely right about having to unpack the terms. If I move forward without collaboration, pretty soon people aren’t familiar with where I am. And I need collaboration to know if where I am actually exists on a map of reality so other people can get there.
If it sounds confusing, blame my father. He was a salesman. He used to say, “if you can’t sell ’em, confuse ’em.”
lambert
if it helps, i am pretty sure i agree with your conclusion. i wish it were more direct and intuitiive. i am not really a pharisee, but i have grown wary of those who if they can’t convince em seek to confuse em.
does not mean i suspect you of this, or even not that your formulation may add something important and precise to our understanding… i’m just not there yet.
Ed –
Your terminology is very counter-intuitive, and that is a barrier to understanding.
To my earlier point, you don’t consider output in your productivity definition, and you do in your effective demand definition.
You might be on to something with your math, but your vocabulary is at odds with common understanding; and quite frankly, that makes it all very confusing.
If you can’t unpack it in a way that makes very clear sense to, frex, Coberly and me, you’ll have an impossibly hard time communicating with a larger population.
Cheers!
JzB
Is effective demand kind of like supply? I mean, if supply is high, but demand is low, will effective demand also be high because there is supply not being utilized?
Critter, I’m thinking Lambert’s “effective demand” is really a kind of POTENTIAL DEMAND. After all, Lambert says “The best way to think about effective demand is the unused capacity of demand in the economy.” What he is describing is demand in an economist’s perfect world, just as Potential output is a measure of real output in that perfect world. Though even that perfect world is far from perfect.
Lambert, your explanations help. But your use of the words “effective demand” troubles me because I think it differs from Keynes’ use of those same words, and from Adam Smith’s words “effectual demand”… both of which refer to the amount of money people actually spend. Effective demand counts my wife’s new Kia but not the Ferrari that I would be driving if I could afford it.
Coberly, I read the first few paragraphs of the post but stopped immediately when I came th this line: “The data for this equation comes from this graph at FRED.”
I went directly to that graph and wrote on scrap paper the two data series names you can see at the very top of the graph: “RCPHBS” and “PRS84006173”
Then I went direct to FRED, typed the first of those names into the search box and discovered that the more meaningful name is “Business Sector: Real Compensation Per Hour (RCPHBS)” here:
http://research.stlouisfed.org/fred2/series/RCPHBS
Then in the search box I typed the second name and found the wordy name to be : “Business Sector: Labor Share (PRS84006173)”
The colons are not operators. They are part of the descriptive names of the data series at FRED.
Lambert, you are doing some interesting stuff.
Oh, yeah,
Lambert, in the comments you said: “Matt’s comment about govt surpluses and trade deficits is important to talk about. It seems government surpluses during a boom time increased private production.”
So, perhaps, it seems. But I challenge the assumption.
The Federal government WAS ABLE TO run surpluses for the same reason the economy was booming. It is because TCMDO debt grew slowly from 1986 to about 1992, making the debt burden relatively small in the early 1990s.
In addition, the quantity of circulating money M1 grew at an unusually rapid rate in the early 1990s, so that people had more money and didn’t need to use so much credit.
The combination of events reduced the “debt per dollar” ratio and reduced the factor cost of money, a cost which competes with wages and profits.
With more money left for productive-sector income, people could afford to buy more, and business didn’t have to price themselves out of the market. The economy was able to grow because people had money.
JzB… Output is in the productivity definition.
Y/L=W/e
And effective demand also has 2 equations…
Effective demand = WL/(labor * capital utilization)
Effective demand = Ye/(labor * capital utilization)
So productivity, labor income, output, real wages, capital income are all entertwined with labor and capital utilization.
Jerry… i have to put words in your mouth to make sense of your question. I assume supply is supply of labor and capital. Thus, if unused supply of labor and capital is high, potential demand is going unfilled. Think of effective demand as potential demand. Yet potential demand is constrained by labor share of income.
The Arthurian… Oh my God! I am responding to the comments one by one, and I just said above Potential demand. You understand!!! YEAH!!!
And you are right about the colon.
Yes, back in the 90’s, people had more money, real wages were rising, The natural rate of interest declined through the 90’s, which would show that people progressively needed more liquidity. Then it jumped higher at the end of 1998, as if all of a sudden there was more liquidity. And the natural rate rose until the 2001 recession.
So I agree with the economy was able to grow, but I place the year at 1998 for the jump in the natural rate. The bubble really started towards the end of 1997. I hope to read your thoughts on this.
Lambert,
The “natural” rate of interest? I don’t understand that. If you have a calculation for it, that would help me. Sounds like you might!
The Arthurian…
Here is a post about the natural rate of interest as calculated by effective demand.
http://effectivedemand.typepad.com/ed/2013/06/the-natural-rate-of-interest-effective-demand-style.html
Lambert
I will drop out. I started a longish “understanding” of what you wrote, and went back to basics quite far. For me this is usually the best way to solve anything. But you have people who are already at least partly on board with you. If you can convince them, that might be best.
If you have trouble convincing others, might be useful to back up and make very clear what you mean. Even here you have a great tendency to assume knowledge that ordinary people just don’t have. Then you throw an equation at them which is hardly transparent.
That is not a criticism, just an observation in case you find it useful.
Coberly,
I appreciate your views. and agree with you. I have been moving forward alone on this. Even alone, I need to move forward.
Of course, it would be wonderful to collaborate on this. It would develop better. But I guess i am laying down some groundwork that others can tweak later.
I have just received an email with data to start looking at Europe. So there is collaboration in the works.
Edward
if you are still around…
see how you are getting along on this. if you have folks who understand you and they and you communicate to the people who matter politically, great.
my observation of economics as discussed publicly is that no one knows what anyone else is talking about, and the economists get lost in their own symbols… “models”. i hope you can avoid this. if you find you are not avoiding it, find some idiot like me and try patiently to explain it to them. hopefully they will ask useful questions.
Coberly,
I am still around because I get the comment feed to my email.
You are absolutely right… people get lost in their models. I am definitely lost in my model because I am discovering new things just about everyday.
Just yesterday I posted what I think is an important part of effective demand… how recessions are understood as real GDP bumping up against the effective demand limit.
http://effectivedemand.typepad.com/ed/2013/07/true-potential-real-gdp-looking-at-previous-recessions.html
Whenever real GDP hits effective demand, utilization of labor and capital screeches to a halt, then there are various changes that can take place, but a recession will follow.
The graphs in that one post could be enough to understand effective demand.