I think some people need to see this graph:

In fact, you can search for the cross correlation between the 10-year %-change in CPI vs monetary base across lead/lag scenarios:

This chart demonstrates that, at peak explanatory power, monetary base offers with a 42-months lead explains less than half of CPI. This means that, even with the strongest correlation found in this lead/lag search, 55.7% of inflation is explained by non-monetary base factors.  And that correlation is sinking every day.

For all of the talk about the necessary mean reversion in corporate profits, let's start with auto-correlation of Corporate Profits as a % of GDP:

No such mean-reversion.  Future profits are actually well correlated with past profits, but the relationship slips into statistical insignificance - but not into negative (mean-reverting) territory to much degree of statistical significance.

Let's take a derivative -- year/year -- to see if _growth_ of profits will demonstrate mean-reversion:

The same story is told here.

Mean-reversion is, at the very least, not supported by historical precedent.

Contributing to the discussion on the nature of growth, I present the lead/lag correlations between population growth and economic growth. 

The correlation between the 30y mortgage rate and 10-year US Treasury Bill rate is generally pretty strong and stable.  From 2004 through the July of 2007, the spread between the two stayed in a tight 54 basis point range.  Since that point, the spread has moved in a much larger 203 basis point range.

With the recent rise in US Treasury yields from their moribund lows, the question of what that will do to the mortgage market has been a popular discussion.

The bottom chart may help:  when the spread is greater or equal to 1.85, the lead/lag correlation curve looks substantially less sure about the relationship between US Treasuries & mortgage rates.

I would posit that UST rates can rise with minimal effect on mortgage rates, but rather serve to narrow the spread.

In our on-going series comparing a single variable across multiple recoveries and recessions, we now present Transfer Payments as % of Personal Income.

Let's take about heading into recession:

Transfer payments as a fraction of income actually increase heading into a recession -- sometimes as early as 2 years before.

What's going on?  How is this possible?

I believe the answer lies in the automatic stablisers -- counter-cyclical spending which occurs without additional lawmaking.  When cyclical forces begin to be a drain on the economy, the blow is softened.  This view seems to be confirmed by the coincident or even leading relationship between Initial Claims Per Population and Transfer Payments as % of Personal Income y/y:

Conversely, let's examine what Transfer Payments as a % of Personal Income looks like coming out recessions:

The present recovery is rather average in this regard -- and looks rather unlike a country headed into recession.

To further explore that thesis, we created an oscillator of Transfer Payments as a % of Income by subtracting its 10-month moving average.  We then applied blue recession bars:

These values are not supportive of any imminent recession, ether.  The only example which saw a recession from below the 0 level was the Fed engineered recession of 1981-1983.

Before a recession begins, auto sales have already begun to flag:

They have also typically lagged recovery in past situations, but certainly not this one:

One thing that looks clear is that it looks nothing like an economy heading into recession.

The evidence continues to mount that imminent recession calls made at the end of the last year were incorrect.  This is something that we were vocally dismissive of throughout last year.

Some historical recovery cycle comparisons to fuel the discussion:

Wage Growth:

Real 5 year rates:

Real Fed Funds Rates:

Just the facts:

• Correlation of 4-wk claims average to payrolls: -0.53

• Correlation of 4-wk claims average to payrolls when claims are < 385k: -0.14

• Correlation of 4-wk claims average to payrolls when claims are > 385k: -0.57

This set of data brings us to a rather solid non-conclusion:  when initial claims are below 385k, they have a particular lack of predictability.

Payrolls also lead post-revision (but less so as released).  See the late 70s:

The late 90s:

And the credit bust:

There are few things which have a stronger leading relationship.  Take quits:

And the very volatile Temporary Professional & Business Service Employees series:

..

The fact is that the month-to-month change in payrolls is one of the earliest indicators of economic activity available, despite it being labelled a coincident indicator.  The problem is that each monthly release is subject to a substantial amount of revision, so much so that it change the tenor of a single report entirely.

As a consequence, we look forward to the payroll numbers, particularly broken down by type (the temporary and business/professional service employment changes perhaps demarcate changes in capital expenditures most quickly), but have to place it in the framework relative to less revision-heavy indicators -- particularly consumption and production.

We've found that comparing apples to apples -- that is, the same points in time relative to the beginning of a recovery or recession -- is a useful technique for giving context to the present.

These types of charts are gaining popularity amongst sell-side research analysts, and are a pain to construct involving a lot of manual work.

We've always found it a treat to view them, so we spent some time this weekend to develop a piece of GNU R code to automatically generate cycle graphs over all of the available data.

Here are three examples.  If you'd like to see any more, let us know, and we can run them.

Let's start with payrolls:

This recovery has been miserable for job growth - which shouldn't be a surprise to anyone.  It however is not the worst post-war employment recovery.  It in fact is outpacing the 2001 recovery (which saw the dot-com bust re-suppressed by the fallout of September the 11th), as well as the Reagan recovery of 1980 -- which was squashed by the Federal Reserve less than a year in.

Initial Claims:

Perhaps surprisingly, the Initial Claims for Unemployment Insurance trend is actually the best of any recovery from 1970-on (when this data first becomes available).  This potentially has a portends a more optimistic future for employment than widely projected.

The S&P 500:

The S&P 500 has experienced the best post-recession performance since index inception.  This should be viewed that it is relative to the magnitude of the largest decline since index inception.∫

Edit 9:45 pm: Adding a few more charts.

Let's finally examine the beginning and end of economic activity - production and consumption.

Retail sales:

Industrial production:

A reminder that upward volatility is volatility, too:

Although the VIX generally doesn't treat as obviously as such, recall that the VIX increased rather substantially from 1996 through the top in 2000.

From 1990-1996, the VIX averaged 13.6.  From 1996-2001, it averaged 18.5 -- despite the S&P 500 increasing 137% during that period of time.

The VIX is a often a great indication of what smart money is doing during divergences, but in our testing it has no value finding tops -- it's of far higher signal value trying to locate bottoms.

That's not a bubble...

That's a bubble.  Those making the recent characterisation of valuations as a second dot-com bubble were not around for the first one.

OpenWave was worth $50B.  Microsoft was worth $486B.  Cisco was worth $425B.  Red Hat was worth $26B.  This accounts for a trillion dollars of market cap -- on $50B of revenue.  And less than $1B of that revenue was for the $76B of market cap of OpenWave and Red Hat.

Japan and USA lead the world:

From the great Econompic blog, Jake continues:

A severe downturn for those without monetary control [..] and surprising strength in Japan [..]

This is a fascinating observation.  These countries all share high advanced-economy style debt-loads, and currency flexibility/monetary sovereignty are very neatly splitting economies across many variables, namely the most important ones:  growth and bond yields.

It is also very good news.  Japan and the USA comprise for 32% of the global economy. If any two economies can help pull the rest of the world into growth, it is these.