PeakTrader.com

[administrator]

http://4.bp.blogspot.com/_otfwl2zc6Qc/SwWyuy39ikI/AAAAAAAAL8g/Cp4XVVbTmqY/s1600/worldgdp.jpg

@@@@Morganovich:

I fear this chart would look a great deal less favorable toward the US if we hadn't changed the way we calculate CPI back in the early 90's.

it's amazing how much systematically understating inflation by 5 or 6% can push up your reported real GDP growth...

the simple fact is that if you believe (as i do) that we have fundamentally distorted the way we track inflation, then most of the rest of even our nominal economic statistics (GDP, productivity, etc) are likewise distorted and anything reported in "real" terms is distorted by definition.

(anything divided by a distorted number = a distorted number)

****PeakTrader:

Morganovich, so, since the early '90s, you believe we've been in a deep depression, the U.S. money market and bond market have been wrong about interest rates (i.e. lenders and investors were too stupid to realize they were losing money), and U.S. living standards declined.

@@@@Morganovich:

peak-

i'd urge you to read this:

http://www.shadowstats.com/article/consumer_price_index

then look at this graph:

http://www.shadowstats.com/charts_republish#cpi

this is not some made up tinfoil hat number. it's simply a comparison between the way were used to measure CPI and how we do it now.

the old way took a basket of goods, priced it at time A then again at time B and used the difference to calculate % change.

the new way takes a basket of goods at time a, prices it at time b, adjusts time b's prices hedonically for improvements etc (but somehow only ever downwards, products are presumed to never get worse, which is clearly untrue in many cases), then geometrically weights it to over-emphasize things declining in price and de-emphasize things that are increasing. if the price of steak goes up, it comes out of the basket and more hamburger goes in. you don't have to believe me, go get the methodology from the BLS.

which of these sound like actual inflation to you?

the latter system will create a downward slope even in a data series that doesn't has no net change. imagine a 2 product consumption basket comprised of A and B. you consume a like amount of each. A increases in price 10%. B declines in price 10%. no net inflation, right?

not according to the BLS post 1992. the BLS will say that you consume more B, less A and that deflation is occurring. amazingly, they could claim deflation even if A had increased 11% in price and the system had clear arithmetic inflation depending upon how they define the elasticities in their geometric weighting.

while price certainly drives demand, demand has already driven price over a given interval of inflation, to go back and reweight again is double counting which may sound sensible, but ultimately makes the data gibberish if you iterate it enough times. run a bounded random walk model with 100 or so commodities over numerous iterations using this kind of geometric system and you'll see the downward bias emerge.

real GDP grew in the 90's. policy was not yet too loose as we were not trying to clean up a bubble and it took a while for the divergence of the map and the terrain to really bite us.

i believe that real GDP growth has never recovered from the 2000 crash and has been negative for much of that time. since then, we have surged from bubble to bubble and run up piles of debt (which is extremely rational if you think inflation is higher than reported)

i think that the systemic misrepresentation of CPI has led to dramatically loose fiscal policy and concatenated asset bubbles that have swamped real growth and skewed the economy substantially and that we are mistaking inflation for growth.
in real terms, we are less well off than in 2000.

it is perhaps worth considering that the reason people complained so bitterly about prices in 2006 is that they were going up so fast.

it is worth considering that the reason so many people feel less well off than in 2000 is because they are.

i realize how preposterous all this sounds. believe me, i do. before dismissing all this as tinfoil hat lunatic ravings, take a serious look at the information. i've tried about 30 ways to poke holes int his thesis, and it pretty much holds water as far as i can tell.

but please, have at it. if you can disprove it, i'd love to see what you come up with.

****PeakTrader:

Morganovich, the GDP Price Deflator (which isn't a fixed basket of goods) implies the CPI is unbiased. I have more faith in markets than articles. There are seasonal, cyclical, and structural factors. Inflation is too much money chasing too few goods. We've just completed a period of excessive consumption, because of declining prices and diminishing marginal utility. There remains too many goods.

@@@@Morganovich:

peak-

how do you figure on GDP deflator?

GDP deflator = nominal gdp/real gdp X 100.

without having CPI, how do you get real GDP?

how does it "imply the CPI is unbiased"?

****PeakTrader:

From article:

"When comparing GDP between years, nominal GDP and real GDP capture different elements of the change. Nominal GDP captures both changes in quantity and changes in prices. Real GDP, on the other hand, captures only changes in quantity."

My comment: Changes in both the GDP Price Deflator and CPI have been similar, although one represents total output, while the other represents a sample. I believe, quality improvements are understated. For example, you can buy a computer worth $1 million 20 years ago for $1,000 today. Also, cell phones didn't exist 20 years ago. Now they do.

@@@@Morganovich:

peak-

i don't think you understand how GDP deflator is estimated.

the weighting argument you make is circular.

also:

quality increases are already built hedonically into CPI. oddly, quality declines never seem to be. my grandmother's blender still works like new after 50 years. a new plastic one breaks every 3-5. why is that not less valuable? a lot of furniture and clothing is similarly flimsier and less durable.

once you get into this hedonic game of quality comparison, where so you stop? it's inherently arbitrary - therefore it should be avoided. it inevitably injects researcher bias into the data.

****PeakTrader:

Morganovich, from AmosWEB Webpedia on GDP Price Deflator:

"There are three important points to note about the GDP price deflator, two pro and one con. These points are especially important when comparing the GDP price deflator to the CPI.

First, the GDP price deflator is based on ALL production in the aggregate economy. This is a definite pro. It includes not just urban consumption, as does the CPI, but also investment expenditures for capital goods, purchases by the government sector, and even exports to the foreign sector. It has it all. So for anyone truly interested in a price index for the aggregate economy, this is it.

Second, the GDP price deflator measures the prices of "current production", the prices of goods actually produced during the current year. This also a definite pro. The CPI, in contrast is based on a market basket of goods identified 5 or 10 years earlier. Again, from a macroeconomic perspective, this is a much better way to go.

Third, the GDP price deflator is reported quarterly, every three months, along with other measures in the National Income and Product Accounts. The GDP price deflator for the first three months of the year (January, February, and March) is not available until late April or May. And then, it is only a 3-month average rather than a month-by-month, blow-by-blow price index. From a timeliness perspective, the GDP price deflator comes up short on providing decision makers with the information needed."

Also, I may add, a high quality blender is much cheaper today in terms of buying power. Consumers can not only afford a high quality blender today (if they choose), but also a microwave that'll last 10 years for under $100, a coffee maker that'll work for years for $20, durable coffee cups for $1, and many other items. Quality improvements aren't fully reflected in the CPI. Autos, for example, are more durable, more efficient, with more features and fewer defects, and cheaper in real dollars. They've improved a great deal since the Model T.

@@@@Morganovich:

peak-

you are confusing the map with the terrain.

GDP deflator is a calculated, modified estimate. it's not a real number. it's not like they count bushels of corn and pair of shoes and blenders. it'd derived, not observed. assuming that GDP deflator is somehow a "real" number is going to lead to errors. it's based on a whole set of estimations, assumptions, and adjustments.

since the 90's, the BLS has shifted the CPI calculation to be more like GDP deflator.

so your argument that one can reinforce the other is totally circular.

all you are saying is that if we use a similar methodology in 2 situations, we get similar results.

the fact is that this methodology is broken.

you can test it with randomized bounded variation series data and see this.

and you're missing the point of the hedonics argument - the point is that it's totally subjective. someone makes this stuff up, then they treat it like it's actual data.

any student of a science, even economics, will tell you that you need to be incredibly careful with data like that. once you introduce researcher bias, your set is pretty much shot.

but the real key to assessing hedonic adjustments is this: "who can judge and how"?

can anyone really claim to know precisely what % a car this year is better (or worse) than last? this problem is far more challenging with goods like clothing or furniture or restaurant meals.

it amazes me that people (peak trader among them) who profess to believe in markets then go on to reject PRICE as a signal of quality in favor of made up, subjective adjustments made by a few bureau/techno-crats.

presuming that a couple of statisticians can more accurately measure year to year changes in the quality of every product in the country at all, much less better than the mass of consumers making individual decisions seems pretty indefensible to me.

****PeakTrader:

Morganovich, you haven't proven the data is biased one way or the other. Yet, you believe inflation is much higher than reported.

It's interesting you believe I contradict myself on markets and reject price when the goods market, money market, and bond market disagree with your theory about inflation. Your implications don't add up, e.g. U.S. GDP contracting at full employment. It's much more likely inflation is overstated than understated, given the rise in living standards.

@@@@Morganovich:

Sure i did peak, you just didn't read the piece. go back to my first response to you and read it.

again, this is not some made up number. it's just a comparison of the old BLS methodology to the new.

i've run the geometric BLS methodology iterative and it creates a downward slant from random price movements in a data set despite the set having no net arithmetic inflation.

since you clearly won't take my word for it, try it yourself. any half decent econometrics modeling package will let you do it in a couple hours.

go get the BLS methodology before and after the boskin/grrenspan changes.

you clearly contradict yourself. hedonics is utterly, irretrievably anti market. it is, by definition, inventing adjustments to the price that emerges from goods markets to attempt to account for real or imagined differences in products based upon the estimations of some economist or bureaucrat.

that's not accepting a market signal, it's twisting one.

the goods market has been increasing in price well ahead CPI if you look at a given basket of goods. that's undeniable.

the bond and money markets are awash in liquidity and trade as you'd expect in such a situation.

your argument doesn't make any sense.

all these markets are behaving precisely as you'd expect if there was too much liquidity and too little growth.

how else to you explain 10%+ gains in gold, bonds, and equities over a 3 month period?

you really believe that we can have rampaging increase in virtually every asset class simultaneously and a constant commodity index up 29% YTD and retail deflation?

you are also confusing nominal and real gdp. you can create jobs in a nominal growth environment. but you'll notice that employment never got back down to the levels of 1999-2000. not that many jobs were created. u-6 dropped from 10% to 8%. only 2 of 3 jobs that disappeared in the 2000 crash came back. have you given any thought to why that might be if growth is so good?

****PeakTrader:

Morganovich, inflation can be overstated or understated using a simplistic and static model that has no meaning. I stated above, there are seasonal, cyclical, and structural factors. There were several major structural breaks, starting around 1980, including the U.S. becoming an open economy, the Information Revolution, the Great Moderation, and the breakdown between money and GDP, which had huge impacts on inflation. The adjustments seem to be unbiased. Anyway, as I stated before, markets, economic fundamentals, and related data all support those adjustments.

You stated: "...you'll notice that employment never got back down to the levels of 1999-2000. not that many jobs were created...only 2 of 3 jobs that disappeared in the 2000 crash came back. have you given any thought to why that might be if growth is so good?"

Yes, because the U.S. economy become more efficient. The U.S. created 17.6 million jobs between 1993-98, and created only 3.7 million jobs between 2001-06. However, U.S. real GDP growth was only slightly higher from 1993-98 than from 2001-06. So, the U.S. became much more productive in the 2000s, i.e. using fewer inputs to produce more output. A structural bear market began in 2000.

The quick and massive U.S. Creative-Destruction process, generally between 2000-02, freed-up resources, e.g. labor, capital, raw materials, energy, etc. Freed-up capital was redeployed into firms that generated even more capital.

@@@@Morganovich:

There was also a huge break in 1992 when we changed the calculation...

and my point is not that you can over or understate inflation with a simple model, my point is that if you plug random data with no arithmetic net change into the BLS methodology, you get a downward trend.

that's pretty much the definition of a downward bias in a methodology.

your real growth comparison only holds if we use the new CPI calculation. using the old one, there was no real growth from 2000-present

and you duck the key question:

you really believe that we can have rampaging increase in virtually every asset class simultaneously and a constant commodity index up 29% YTD and retail deflation?

seriously, think back to 1999. prices of everything were rampaging. we had some of the lowest unemployment on record. you really think inflation was 2.6% that year? 1.6% the year prior? could it be that the massive failure of the short run phillips curve to function was not because we entered a magic land of productivity and overcapacity but because we were miscalculating inflation?

****PeakTrader:

Morganovich, your model doesn't explain anything. It's a static model that shows accelerating inflation using obsolete information.

Your own statement proves the old method is obsolete: "Your real growth comparison only holds if we use the new CPI calculation. using the old one, there was no real growth from 2000-present."

Do you really believe, U.S. real GDP growth was zero at full employment? Also, why would anyone lend or invest at 5% when inflation is higher than 5%?, and why would foreigners invest in a stagnant or contracting economy?

When there's excessive economic slack, there's deflation. That's the case today. A higher level of productivity is disinflationary, which was the case beginning in the mid-'90s. Imports are also disinflationary when prices fall to induce demand, and real wages rose in the second half of the '90s.

Gold seems to be replacing the dollar as the world's reserve currency, since the dollar has depreciated, and it's uncertain whether or not the Fed can eventually drain enough dollars to preempt inflation. Commodity prices rose, because developing countries are growing faster. They're producing heavier goods, unlike the U.S. economy.

I believe, what ended the first superbubble from 1995-00 were budget surpluses or contractionary fiscal policy (mostly through more taxes as the economy expanded), the end of an 18-year bull market much of it based on demographics, and a correcting mechanism to keep future labor supply and demand in equilibrium (because the economy cannot support too many workers retiring early). It was remarkable, the U.S. had a second superbubble, in a structural bear market, from 2002-07, after the Creative-Destruction process, which not only made high-tech firms more efficient, but also old economy firms, although the second superbubble was based more on consumption than production.

Moreover, I may add, there's no way there can be zero real growth in the middle of a homebuilding boom. Housing starts and new home sales can be counted (and larger houses were built based on square feet). Furthermore, domestic auto units produced can be counted (and roughly half were SUVs, light trucks, and minivans, rather than smaller autos). New homeowners tend to buy furniture and appliances for their homes, and there was also over $300 billion a year in home improvements during much of the 2000s. Units and dollars can be measured.

@@@@Morganovich:

peak-

you're really not getting this so i'll put it as simply as i can:

take the current BLS methodology for CPI.

instead of price data, use random price fluctuations that have a zero average (which is to say 0% inflation using an arithmetic measure).

run the model 30 times iteratively.

you will get a downward slope in CPI. this is the nature of automatically overwieghting things that decline in price and underweighting those that rise.

this tells you the model is biased to the downside.

this means the current CPI measure understates inflation.

by how much is an interesting and open ended question, as the simple fact is that all inflation measures are estimates.

you can bluster all the pop econ books you want, but the fact is that the CPI model is demonstrably broken.

you think the 2000 bubble burst due to budget surpluses? that's so bizarre i don't even know where to start. you really are a bubble baby.

if you think housing drives the economy, you haven't been watching healthcare and services industries. they utterly swamp home building. furniture comes from overseas. so do most appliances etc.

lending decisions are driven as much by cost of capital and access to leverage as inflation.

if i can borrow at 25 bp, i'll happily lend at 5% if i can do it with leverage. the current situation of borrow at 25 bp and buy us bonds at 10 or 15 to 1 makes for a pretty tidy profit at a money center bank, no?

notice that commercial lending still sucks for all but the largest companies? this is because you can't lever up those loans the same way.

further, what do you think drove the insane borrowing bubble here in the US if not negative real interest rates? the whole mortgage market thrived on it.

we have repeatedly tried to cure the first bubble by inflating new ones. you seem surprised that they keep occurring, but so long as we keep monetary policy this loose, nothing's going to stop them.

****PeakTrader:

Morganovich, inflation reflects exponential growth (a 3% rise in inflation is a rise from the prior year). So, the arithmetic mean would overstate inflation. You can see this using a compound annual growth rate calculator. Type 140 for ending value, 100 for beginning value, and 4 for number of periods. The answer is 8.78% (not 10%). The more periods, the lower the rate, e.g. 180, 100, and 8 is 7.62% (not 10%).

&&&&Steve:

Here's one simple fact that any off-the-street reader (i.e. me) will agree with:

Peak Trader consistently refuses to address morganovich's central argument.

The reverse is not true.

This doesn't mean morganovich is right. But it's damned suggestive.

Steve
http://asymptosis.com

Complete Debate:

https://www.blogger.com/comment.g?blogID=28997633&postID=4523660180183085255

-------------------------------------------------------------------------------

http://4.bp.blogspot.com/_otfwl2zc6Qc/SwWyuy39ikI/AAAAAAAAL8g/Cp4XVVbTmqY/s1600/worldgdp.jpg

Article:

"Somewhat surprisingly, the Economic Research Service of the U.S. Department of Agriculture has some great international historical macroeconomic datasets. According to its website:

The International Macroeconomic Data Set provides data from 1969 through 2020 for real (adjusted for inflation) gross domestic product (GDP), population, real exchange rates, and other variables for the 190 countries and 34 regions that are most important for U.S. agricultural trade.

The chart above shows the annual shares of real world GDP for four geographical regions (European Union 15, Asia/Oceania, Latin America and the combined share of Africa and the Middle East) compared to the U.S. share of world GDP between 1969 and 2009 (data here). What might be surprising is that the U.S. share of world GDP has been relatively constant for the last 40 years, and is actually slightly higher in 2009 (26.7%) that it was in 1975 (26.3%). It's also interesting that the EU15's share of world GDP has declined from about 36% of world output in 1969 to only 27% in 2009. Further, despite having a large share of the world's oil reserves, the Middle East's share of global output has increased from only 2.23% in 1969 to 3.16% in 2009 (graph shows Middle East combined with Africa).

Bottom Line: World GDP (real) doubled between 1969 and 1990, and has increased by another 60% since then, so that world output in 2009 is more than three times greater than in 1969. We might mistakenly assume that the significant economic growth over the last 40 years in China, India and Brazil has somehow come "at the expense of economic growth in the U.S." (based on the "fixed pie fallacy") but the data suggest otherwise. Because of advances in technology, innovation, and significant improvements in U.S. productivity, America's share of total world output has remained remarkably constant at a little more than 25%, despite the significant increases in output around the world, especially in Asia.

Update: The chart above represents about 91% of the world economy and does not include Canada and the European countries not included in the EU-15."