Credit inflation, Deflation: Prechter Interview

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

"Interesting" too (in the sense of the old Chinese curse about living in interesting times ;)) that the house price appreciation hasn't had nearly as much effect on the net worth of the average household as almost all the news would have one think - and what an understatment that is as shown by the data below derived from Z1 and Census data.

Some raw numbers (1st qtr 2000 was the highest value in the last 10 years):

Date--------Net worth/Household-----with CPI+lies correction (base 100 = 1975)
1/2000---------$217,681-------------------$56,885
4/2006----------165,047---------------------34,936

There are roughly 2.7 people per household per the Census bureau for what its worth.

Sure it hasn’t - people have turned around and borrowed that "house price appreciation"! And as you suggested before, the key is real net worth.

But before we even look at the statistics, we should ask ourselves why in the world should average household wealth have increased? What could have made those houses really worth more? After all, they provided the same shelter and amenities after a 50% "price appreciation" that they did before. Since when did printing money and debasing the currency ever create wealth?

What really happened is the Fed created the illusion of greater wealth, prompting households to borrow against it and send it to Asia and OPEC - if you want to see increased wealth, all you need do is look there.

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

Now that I know what the FDR is…

Tsk

tsk

tsk.

You haven’t been keeping up with the FFF…

, its pattern reminded me of one of my own charts. This one adds up all the various Fed actions I track - interesting how it peaked around 2003 around 15% and also has other broad similarities to the FDR.

As it should. After all, we are both attempting to measure inflation, via independent means. Taking the rate of change of the FDI and plotting the resulting rate of inflation probably makes the connection more clear.

In your estimation, by how much does the FDI lead CPI?

In very rough terms, the CPI is like taking a four year trailing moving average. Mathematically, the net delay is therefore about two years. But the effects are distributed - smeared out - some show up pretty quickly, others in as many as four or more years. Plus the delay varies somewhat depending on the state of the cycle.

Would it be correct to assume that the FDI attempts to measure *all* inflation, not just consumer prices, but also asset price changes?

Yes. Consumer prices would be adequate if all we bought with and sold for dollars were consumer goods and services. But we don’t. If you wanted to observe the total effect of changes in the value of the dollar, you’d have to take into account all those things against which dollars are exchanged.

The failure of a consumer price index alone to do this is the source of much mischief and misery. During the 1990s, the financial and economic community assumed that because consumer price inflation remained benign, the Fed's copious monetary and credit expansion was not having an inflation effect. Greenspan erroneously chalked it up to "productivity". It was inflationary, however; it was just that the excess money was largely being directed into the financial markets. It was sort of being "stored" there; and was only a matter of time before it spread out to drive up the prices of other things like barrels of oil and such.

Another important point is that prices in financial markets - those for things like stocks, bonds, and commodities - change in real time. They’re repriced by the minute. Consumer prices and wages aren’t. That is why a true index of broad inflation responds to inflationary transients in a much more timely way.

Do you intend it to also measure US inflation on a global basis, in other words, also including dollar purchasing power gain or loss?

Absolutely. If the Fed dumps a load of money into the system, you don’t see domestic consumer prices jump up right way. But the value of the dollar in forex markets can (and often does) reflect the inflationary action quickly. Forex markets even anticipate such things, for example when the dollar falls in the wake of weak US economic data. The fall is largely due to markets anticipating the Fed will become more inflationary.

Think of it this way. The inflationary act does little or nothing to domestic wages and prices right away. But wages in, say Germany or Japan - as measured in US dollars via the exchange rate mechanism - can rise almost instantly. So the FDI is sort of like taking the conventional dollar index, adding a component for domestic wages and prices, and then adjusting each of the other currencies in the basket for domestic wages and prices in its own economy. A "real" USDX, so to speak; one that that should generally track the conventional forex-based USDX but superimposed upon a persistent down trend as the other currencies in the basket themselves are inflated.

[bart]

Re: Credit inflation, Deflation: Prechter Interview

Tsk

tsk

tsk.

You haven’t been keeping up with the FFF…

I'll flagellate myself a few extra times tonight to help make amends for such a heinous oversight.
All hail non transparent TLAs from the land of the Finster? ;)

As it should. After all, we are both attempting to measure inflation, via independent means. Taking the rate of change of the FDI and plotting the resulting rate of inflation probably makes the connection more clear.

Or, you could back out your rate of change to match my chart? :p ;)

In very rough terms, the CPI is like taking a four year trailing moving average. Mathematically, the net delay is therefore about two years. But the effects are distributed - smeared out - some show up pretty quickly, others in as many as four or more years. Plus the delay varies somewhat depending on the state of the cycle.

I'm way more familiar with the varying lags than I'd like to be, like in my inflation prediction work... *sigh*
Thanks for the +/- 2 year estimation.

Are you considering drawing another line of the chart to show estimated future inflation, considering that ball park 2 year lag? Or is it just too dicey considering the varying lags?

To what cycle are you referring - the 4 year business cycle?

Yes. Consumer prices would be adequate if all we bought with and sold for dollars were consumer goods and services. But we don’t. If you wanted to observe the total effect of changes in the value of the dollar, you’d have to take into account all those things against which dollars are exchanged.

The failure of a consumer price index alone to do this is the source of much mischief and misery. During the 1990s, the financial and economic community assumed that because consumer price inflation remained benign, the Fed's copious monetary and credit expansion was not having an inflation effect. Greenspan erroneously chalked it up to "productivity". It was inflationary, however; it was just that the excess money was largely being directed into the financial markets. It was sort of being "stored" there; and was only a matter of time before it spread out to drive up the prices of other things like barrels of oil and such.

Another important point is that prices in financial markets - those for things like stocks, bonds, and commodities - change in real time. They’re repriced by the minute. Consumer prices and wages aren’t. That is why a true index of broad inflation responds to inflationary transients in a much more timely way.

Ah yes, grasshopper - the wonderful world of the lag and how many ways it can vary. Aren't they just bundles of fun (not!).

I had a breakthrough of sorts about a year ago when I realized that even the same input can have multiple lags, depending on who is viewing the stat and what their purpose is.

Absolutely. If the Fed dumps a load of money into the system, you don’t see domestic consumer prices jump up right way. But the value of the dollar in forex markets can (and often does) reflect the inflationary action quickly. Forex markets even anticipate such things, for example when the dollar falls in the wake of weak US economic data. The fall is largely due to markets anticipating the Fed will become more inflationary.

Think of it this way. The inflationary act does little or nothing to domestic wages and prices right away. But wages in, say Germany or Japan - as measured in US dollars via the exchange rate mechanism - can rise almost instantly. So the FDI is sort of like taking the conventional dollar index, adding a component for domestic wages and prices, and then adjusting each of the other currencies in the basket for domestic wages and prices in its own economy. A "real" USDX, so to speak; one that that should generally track the conventional forex-based USDX but superimposed upon a persistent down trend as the other currencies in the basket themselves are inflated.

So fast acting inputs like the dollar index (whether you actually use it or not) are what takes care of classic money supply issues like velocity and its potentially quick changes?

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

I'll flagellate myself a few extra times tonight to help make amends for such a heinous oversight.

Awright, this is a family forum here…

All hail non transparent TLAs from the land of the Finster? ;)

Might if I knew what the heck it was…

Or, you could back out your rate of change to match my chart? :p ;)

Nah, I like it with the flaws in your chart there for all to see…

I'm way more familiar with the varying lags than I'd like to be, like in my inflation prediction work... *sigh*
Thanks for the +/- 2 year estimation.

Ditto that *sigh*…

;)

Are you considering drawing another line of the chart to show estimated future inflation, considering that ball park 2 year lag? Or is it just too dicey considering the varying lags?

It's important to keep in mind that "future inflation" isn't what the FDI measures. I can’t emphasize strongly enough that the FDI only measures current inflation. It is the CPI which measures past inflation. I suppose I could try and ‘predict’ the CPI, but why go through the exercise of ‘predicting’ tomorrow’s recap of what happened yesterday?

But before you point it out first ( :p ), it should be admitted that the FDR (the FDI rate of inflation, not the President) does involve a bit of forecasting. A rate of change calculation inherently involves differences between points. Today’s FDI, for example, by itself tells you nothing about the rate of inflation. To do that, you have to compare today’s FDI with the FDI at some other time, for example, the FDI one year ago. When you do that, you can say the dollar lost some amount of value over the past year, and therefore that the average rate of inflation over the past year was X.

It’s a lot like checking your watch when you leave for the grocery store and then checking it again when you get there, and dividing the distance traveled by the amount of time it took to get your average rate of speed. Merely recording where you were at a single point in time alone tells you nothing about how fast you went.

So if comparing the current FDI with that of one year ago gives us the average rate of inflation over the past year, how do we know what the rate is today? Well we could use a closer past point, say six months ago, along with today’s FDI and say that inflation averaged Y over the past six months. That gets us closer to the rate today. We could use progressively shorter time periods and keep getting closer, until were using the FDI as of a few seconds ago versus now’s FDI, and we could say that was essentially the current rate of inflation. But that amplifies short-term variations, including any noise in the data, so what you’d wind up with would be a very spiky and useless chart. So I use a longer time period in conjunction with a smoothing filter.

This means that for the most current inflation rate, I have to use some predictions of near term future FDI values. The filter I use is actually a two year triangle differentiator. That means that the inflation rate shown in the FDR is actually based purely on past FDI values up to one year prior. Slightly later, the FDR value is based mostly on actual past FDI values and a little bit on forecasted FDI values. The current rate of inflation is based half on historical FDI values and half on forecasted values. (Those values, by the way, are the ones shown in my forecast page.) Only after the passage of another year will that value be purely historical.

This is no different than any other calculation of the rate of inflation, by the way. You can only get a CPI rate of inflation by comparing CPI readings, e.g. last month’s reading with that of a year ago or a month ago. There is no CPI datum that says "the rate of inflation is now ___".

To what cycle are you referring - the 4 year business cycle?

Both that and the secular cycle in which it is embedded. Guess I should have said "cycles"…

Ah yes, grasshopper - the wonderful world of the lag and how many ways it can vary. Aren't they just bundles of fun (not!).

I had a breakthrough of sorts about a year ago when I realized that even the same input can have multiple lags, depending on who is viewing the stat and what their purpose is.

So fast acting inputs like the dollar index (whether you actually use it or not) are what takes care of classic money supply issues like velocity and its potentially quick changes?

You could certainly look at it that way. The idea is to have an index of the market value of the dollar. The CPI (however flawed) is one. The USDX is one. But the USDX and CPI differ in that only for the USDX do we have real time data. In that respect, the FDI is more like the USDX. The main difference is the USDX tacitly assumes that the value of the other currencies in the basket against which the dollar is measured do not themselves change in real value. But virtually all currencies are subject to the loss of value through inflation. Hence my analogy suggesting the FDI is like a USDX except where the component currencies themselves are adjusted for inflation.

[jk]

Re: Credit inflation, Deflation: Prechter Interview

Quote:

Originally Posted by bart Are you considering drawing another line of the chart to show estimated future inflation, considering that ball park 2 year lag? Or is it just too dicey considering the varying lags?

I might if I thought that "future inflation" was what the FDI measured. I can’t emphasize strongly enough that the FDI only measures current inflation. It is the CPI which measures past inflation. I suppose I could try and ‘predict’ the CPI, but why go through the exercise of ‘predicting’ tomorrow’s recap of what happened yesterday?

iirc the ratio of coincident to lagging indicators makes a pretty good leading indicator for the economy. i'm curious if you've looked at something like that.

[bart]

Re: Credit inflation, Deflation: Prechter Interview

iirc the ratio of coincident to lagging indicators makes a pretty good leading indicator for the economy. i'm curious if you've looked at something like that.

It sure does seem to me that he should - if for no other reason than to test one of more theories on how it might do a decent prediction job.

The absolute rate could be significantly off, but changes in trends could be quite helpful in timing. Its the primary reason I did my own attempt at CPI prediction.

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

iirc the ratio of coincident to lagging indicators makes a pretty good leading indicator for the economy. i'm curious if you've looked at something like that.

No disagreement on that point, jk. But for my work measuring "the economy" just isn't very helpful. It would be like measuring love or anger. I need hard, definable, objective data, preferably not susceptible to statistical massage.

The conventional leading, lagging, and coincident indicators IMO are most useful for forming a subjective impression of the level of economic activity and an independent assessment of inflationary bias. But basically most of what passes for a forecast of inflation is really a forecast of inflation data. Data which themselves lag, so the construction of a "forecast" really amounts more to merely trying to strip out the lag.

This, for what it's worth, is IMO what the Fed is talking about when it speaks in terms of "pre-emptive" policy action. Despite the actual (lagging) inflation readings, those guys are smart enough to sense when inflation is changing in advance of when those changes finally make their presence known in the inflation statistics. Probably the recent spate of Fed speak reflects their growing suspicion that the tightening so far this cycle has not yet been sufficient.

[bart]

Re: Credit inflation, Deflation: Prechter Interview

Awright, this is a family forum here…

*mumble, mumble*... EJ let me get away with this image as an alternate characterization of iTulip... ;)

Might if I knew what the heck it was…

Just another attempt at getting an "even". TLA stands for Three Letter Acronym - like FDR and FDI... and both are undefined either in context or on the charts emanating from Finster-ville... ;)

Nah, I like it with the flaws in your chart there for all to see…

That'll get you an another rimshot for your collection:
http://www.nowandfutures.com/grins/rimshot.mp3

...
This is no different than any other calculation of the rate of inflation, by the way. You can only get a CPI rate of inflation by comparing CPI readings, e.g. last month’s reading with that of a year ago or a month ago. There is no CPI datum that says "the rate of inflation is now ___".

I do actually attempt that with my inflation prediction chart by fiddling with various "money" numbers and applying lags but given the BLS "creativity" in calculating the CPI, its only accidental if I get close in any given month.

Both that and the secular cycle in which it is embedded. Guess I should have said "cycles"…

Now I'm confused again - the secular cycle in what? The FDI itself?

Don't go telling me you're channeling Kondratieff now... ;)

You could certainly look at it that way. The idea is to have an index of the market value of the dollar. The CPI (however flawed) is one. The USDX is one. But the USDX and CPI differ in that only for the USDX do we have real time data. In that respect, the FDI is more like the USDX. The main difference is the USDX tacitly assumes that the value of the other currencies in the basket against which the dollar is measured do not themselves change in real value. But virtually all currencies are subject to the loss of value through inflation. Hence my analogy suggesting the FDI is like a USDX except where the component currencies themselves are adjusted for inflation.

Yes, I do (mostly and I hope) get that the FDI measures current inflation... but by virtue of everything else lagging, it almost by default must lead - if in no other area than sentiment and perception.

True too that the USDX has the built in flaw of only comparing against other variable value fiat currencies, and that the FDI is one of the few (if not the only) effort that recognizes that fact and tries to offer a better measure. I guess the real issue is that I don't know why you don't or haven't yet done anything in the prediction area...

And of course, I do have private and non published charts, and work in progress... maybe that's it?

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

*mumble, mumble*... EJ let me get away with this image as an alternate characterization of iTulip... ;)

…

Just another attempt at getting an "even". TLA stands for Three Letter Acronym - like FDR and FDI... and both are undefined either in context or on the charts emanating from Finster-ville... ;)

…

That'll get you an another rimshot for your collection:
http://www.nowandfutures.com/grins/rimshot.mp3

Oh you are in top form tonite…

I do actually attempt that with my inflation prediction chart by fiddling with various "money" numbers and applying lags but given the BLS "creativity" in calculating the CPI, its only accidental if I get close in any given month.

You grok. In fact, didn’t we have a discussion about what your inflation prediction chart was predicting? Correct me if I’m wrong, but I think we concluded that it was actual inflation, not the CPI per se, that was the bogey.

Precisely the same issue crops up with attempting to use the FDI as a CPI forecaster. There is so much statistical massage in the CPI that it is almost arbitrary.

Now I'm confused again - the secular cycle in what? The FDI itself?

Don't go telling me you're channeling Kondratieff now... ;)

That 32-36 year cycle of hard and financial assets. But really, the key point was that the lag varies, not so much according to any regular cycle but more according to the composition of the price sets that are undergoing the most change.

True too that the USDX has the built in flaw of only comparing against other variable value fiat currencies, and that the FDI is one of the few (if not the only) effort that recognizes that fact and tries to offer a better measure. I guess the real issue is that I don't know why you don't or haven't yet done anything in the prediction area...

It’s not even so much that it is a "flaw" in the absolute sense. Understood as a relative measure between currencies, it does exactly what it purports to. Only if it is assumed be "the dollar" without such qualification is it really flawed, and then the flaw is really in the interpretation rather than the measure itself.

As for the prediction issue, see again my discussion above about the FDR. Also, check out my forecast page, which among other things, does actually plot a forecast of the FDI out a year (bottom chart). FWIW, the current FDI forecast shows a small deflationary wave in the second half of next year.

Yes, I do (mostly and I hope) get that the FDI measures current inflation... but by virtue of everything else lagging, it almost by default must lead - if in no other area than sentiment and perception.

And of course, I do have private and non published charts, and work in progress... maybe that's it?

Maybe I am being too pedantic about the forecasting issue, but it is mostly to drive home the point that the CPI does not merely understate inflation, but lags it as well. It is off not only in the amplitude domain but the time domain as well. This is exceedingly important, as it has repeated led to policy errors as I alluded to above.

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

This, by the way, is a comparison of the USDX - both the FRB and SGS (Williams) versions - with the FDI from 1985 to present, with the FDI scaled to 1985=100 as with the USDX. Note both the general similarity as well as the greater overall dollar decline in the FDI as compared to the USDX. The latter reflects the absolute decline (inflation) in the base currencies composing the USDX basket.

[jk]

Re: Credit inflation, Deflation: Prechter Interview

this discussion raises a question i've been pondering: are currencies a good hedge or investment? it is hard to see how we avoid a severe depreciation in the value of the dollar, however measured. but surely other currencies are going down in absolute value [however measured] as well, just more slowly.

so, investment questions: is there a scenario in which one is better off investing in currencies instead of in commodities? is there a scenario in which one is better off investing in industrial/agricultural commodities instead of in precious metals?

i am eager to hear thoughts on these questions, because they've been rattling around in my mind for quite some time.

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

this discussion raises a question i've been pondering: are currencies a good hedge or investment? it is hard to see how we avoid a severe depreciation in the value of the dollar, however measured. but surely other currencies are going down in absolute value [however measured] as well, just more slowly.

so, investment questions: is there a scenario in which one is better off investing in currencies instead of in commodities? is there a scenario in which one is better off investing in industrial/agricultural commodities instead of in precious metals?

i am eager to hear thoughts on these questions, because they've been rattling around in my mind for quite some time.

They can be excellent if you're running a hedge fund or money center bank trading operation. Most of the rest of us are IMO better off not bothering with it. Okay, maybe if you have a very large portfolio and spend a lot of time on research, but for my money there are much better alternatives.

Most investors are probably best off only concerning themselves with two forms of money - hard money and their coin of the realm (dollars, for us Americans). To the extent we need a hedge against the value of our dollars falling, hard money commodities like gold, silver, platinum and copper will do everything we need them to without our having to become forex experts. Not only that, but as you allude, foreign currencies themselves inflate and decline in value over time, so in the investment race it's kind of like betting which horses will lose least ignominiously.

It’s probably best to regard hard money and the rest of the commodity complex as effectively different asset classes. You might, say, have a sort of ‘doomsday’ cache of coins and bars as a hedge against very bad times, and it will appreciate at least in nominal terms even if there’s only garden-variety inflation. At the same time, you might dedicate a portion of your portfolio to commodity futures. Specialists like Bart prefer to do so directly. Others, such as myself, prefer to use some of the new ETFs that have been created for this purpose. One, DBC, is linked to the Deutsche Bank Commodity Index. Another, GSG, tracks the Goldman Sachs Commodity Total Return Index. Two others - GSP & DJP - are actually notes traded on the NYSE linked to the GSCI and DJ-AIG commodity indices. They all include exposure to agricultural and industrial commodities. Then there’s CEF, which is a closed-end exchange-traded fund whose assets are basically just gold and silver bullion.

Also remember that stocks - even US stocks - are still a fundamentally different asset class from currency. So if you want to hedge against a declining dollar, stocks are a viable option. By using a mix of stock ETFs and commodity ETFs as well as hard money, you can be just about as currency hedged as you like without the vagaries of the forex markets.

[jk]

Re: Credit inflation, Deflation: Prechter Interview

i think the prize you're looking for with respect to currencies is "least outstanding in an ugly contest."

i've certainly seen the currencies diverge from commodities and from pm's, but those are possibly purely short term phenomena. any thoughts on whether there is a diversification/volatility benefit to using currencies AND commodities AND pm's?

[Finster]

Re: Credit inflation, Deflation: Prechter Interview

i think the prize you're looking for with respect to currencies is "least outstanding in an ugly contest."

i've certainly seen the currencies diverge from commodities and from pm's, but those are possibly purely short term phenomena. any thoughts on whether there is a diversification/volatility benefit to using currencies AND commodities AND pm's?

Maybe someone else will pop in and take an opposing view, jk, but I sure can’t think of any meaningful extra benefit. As alluded to before, possibly someone with a very large portfolio and who likes to spend a lot of time on its management might prefer it, but even then any advantage would come from management skill and effort, which could just as easily cut the other way.

It’s practically a truism of investing that the single most important determinant of risk and return comes not from the selection of individual assets, but from the mix of asset classes. Currencies, including the US dollar, are all currencies. The major diversification benefit would come not from the inclusion of other members of the same asset class, but from the addition of other asset classes, like stocks, bonds, real estate, commodities and hard money.

Let’s take a look at an example $320,000 portfolio. For good measure, you could throw in some coins, cash in the bank (or under the mattress), and maybe some real estate and energy trusts, but just to keep it simple, let’s have it composed entirely of ETFs.

$40,000 SHY (1-3 year Treasuries)
$40,000 TLT (20+ year Treasuries)
$20,000 DJP (Commodity index)
$20,000 GSP (Commodity index)
$20,000 DBC (Commodity index)
$20,000 CEF (Gold & Silver)
$50,000 SPY (Large-Cap US Stocks)
$40,000 VXF (Small-Cap US Stocks)
$40,000 EFA (Foreign Developed Market Stocks)
$30,000 EEM (Foreign Emerging Market Stocks)

This is pretty much a sleep-like-a-baby portfolio. You could turn off the news and forget about it for a couple years at a time. Particularly if you are worried about a decline in the value of the dollar. You have 25% in short and long term bonds, 25% in commodities and hard money, and 50% in global stocks. The first two positions would likely decline in real terms, after all, they are dollar-derivatives. But what about the next four? Combined they represent gold, silver, oil, gas, wheat, corn, and sundry other commodities. With a diminished and diminishing dollar, you don’t really think it’s going to cost less dollars to buy these things, right? Quite the opposite. It’s going to take more dollars to buy them; their prices would soar. If they fell, we’d be talking about a rising dollar (deflation), not a falling dollar (and then the first two positions would rise).

Between the four stock ETFs, you own the world. Now if the dollar declines, as measured against what did it do so? All the publicly traded equity capital in the world? If that capital fell in dollar terms, you’d have at least half of a bullet-proof case that the dollar actually rose in value, not fell. If, perhaps due to some global communist revolution, the world’s capital stock actually declined in real value, there’s not a whole lot left you could say it declined against except for commodities. This would take us back to our commodity positions. All in all, its hard to imagine any foreseeable scenario where such a basket of assets would lose intrinsic value over any realistic investment time frame, and with a little insight and effort, ought to actually grow your real purchasing power come what may.

[jk]

Re: Credit inflation, Deflation: Prechter Interview

it's interesting to me that you think of the dollar and all the other currencies as in the same asset class - "currencies." makes sense. but in my mind i group together foreign currencies, commodities and precious metals - "anti-dollars." but you're right to remind me that it's the overall asset allocation that counts the most, and it may not be worth the time to worry about allocation among my anti-dollar subcategories.

btw, my sleep-like-a-baby portfolio has no net exposure to equities at present. the major equity exposure is via the hussman fund, which is fully hedged at the moment. the minor equity exposure is in canadian trusts which are mostly energy, and also hedged with shorts.