Robert H. Heath

Jobless Recoveries Are a Recent Phenomenon

The "jobless recovery" phenomenon dates back to the 1990 recession. Midway through Clinton's first term in late 1994, early 1995, commentators were still referring to the jobless recovery (and blaming it on President Bush... some things never change.)

This chart (total employment, not seasonally adjusted, establishment survey) clearly shows that employment levels, which in earlier recoveries rebounded quickly at recession's end, recovered more gradually coming out of the 1990 and 2001 recessions.

research.stlouisfed.or...

Surely, in the post-war recessions up to and including the early 80's double-dip recession, employment levels during recessions were driven by contraction and expansion of the manufacturing sector, which responded quickly with layoffs and re-hiring as the economic outlook changed. Since then, employment changes have been more moderated.

The following chart (manufacturing payrolls, seasonally adjusted this time) shows that in the 1990 and 2001 recessions, manufacturing payrolls actually shrunk after the recession's end (consistent with Tom E.'s comment that firms use recessions to off-shore manufacturing jobs).

research.stlouisfed.or...

But it's too simplistic to ascribe the change in post-recession employment too off-shoring of manufacturing jobs. The second chart shows that manufacturing jobs have declined by a little over 2 million since their peak just before the current recession. With a labor force of about 140 million, the direct result of the decrease in manufacturing jobs during the current recession is an increase in the unemployment rate of about 1.4%, only about one-quarter of the 5.3% increase in the unemployment rate during the recession. So the bulk of the increase in unemployment has come from non-manufacturing jobs.

Indeed, it seems that non-manufacturing jobs may have more "flex" in per-capita productivity, so that as the economy expands, firms can expand output with a smaller number of workers (for awhile) before they are compelled to begin hiring again.

Buffett's BNI Purchase: Bearish Bet on the Economy?

How does an investment in a purely domestic railroad provide " the perfect opportunity to diversify out of the US dollar?"

I'm scratching my head over this one.

Why Apple Is Worth $80

Clarification, if g = r, PV is infinite... if g > r, the equation is meaningless.

Why Apple Is Worth $80

I wanted to keep my post(s) short so I've broken them into two different posts.

You write,

"In this example, it assumes that AAPL will grow at 18.8% FOREVER, and that gives the target of $209."

This is wrong... If the Bloomberg model assumed AAPL will grow at 18.8% forever, the "Theoretical Price" would be infinite (with the given discount rate assumptions).

Remember that the DDM model (in simplified form) says that present value of the stock price equals the discounted value of the perpetual dividend stream, or

PV = D / (r - g)

where
D = next period's dividend per share
r = discount rate
g = dividend growth rate

If your dividend growth rate (g) is greater then your discount rate (r), then the PV (the stock price prediction) is infinite. Look at your DDM model on Bloomberg. Your discount rate is 9.671% (3.391% + 6.290%).

You're right that inputs matter a lot when using valuation models, but understanding the underlying math is pretty important too.

Why Apple Is Worth $80

Boy, any post on AAPL never fails to bring out the fanboys and the bashers. It's amazing how much emotion this stock brings out in its admirers/detractors. But let's focus on facts.

I have no position in AAPL, but it's important to point out an apparent flaw in your use of the Bloomberg Dividend Discount Model (DDM) function. The Bloomberg DDM has two inputs for growth rate. The first, perhaps misleadingly called "Long Term Growth Rate" (on the left) appears to be the dividend growth rate over a specified, finite period of time ("Growth Years"). The model then uses another input ("Transitional Years") to smooth the dividend growth rate down to "Growth Rate at Maturity" (on the right) which in both your runs seems a perfectly reasonable 5.319% in nominal terms, roughly equal to long-run (nominal) GDP growth.

I should add that in the absence of dividends (as is the case with AAPL) the model appears to assume growth of *earnings* according to the user inputs until the "Payout At Maturity" or payout ratio assumption kicks in.

Your run of the model with "Long Term Growth Rate" set to 5.000% is assuming that AAPL's near-term EPS growth rate is 5.0%, massively below recent results and consensus estimates.

You should probably re-think your price target.

How Paulson Gave Goldman the Lehman Heads-Up

Has everyone already forgotten what was going on in September 2008?

Go back to any news story from that month. The big problem was that no one -- not the banks themselves, not Hank Paulson, not Ben Bernanke --had any idea what liquidation value could be placed on a huge pile of "troubled assets". But it was abundantly clear that in a distress sale, the liquidation values would be a small fraction not only of face value, but of the value the assets might fetch in a less distressed market environment. For this last reason, among others, it was also pretty unlikely that any firm was going to voluntarily say, "We looked at our assets and they're so impaired you might as well shut us down now."

So, if you're Treasury secretary and you're trying to figure out whether a particular bank is worth saving, doesn't it make sense to have an unrelated party take a look at the assets to draw some conclusion about whether there's value that exceeds the liabilities before you decide whether to throw the tax-payers' money at it? It's called due diligence. And since Treasury didn't have the staff to do the analysis, they needed someone else to do it. Goldman's pretty good at that sort of thing.
Not much scandal here.

Memo to Newspapers: Content Doesn’t Matter Without the Package

"Newspapers’ inability to generate the same revenue online as in print has nothing to do with content. It’s because on the web they are no longer in the business of packaging content, and that’s what the newspaper business, like every other media business, has always been about."

Exactly right.

You might find the following interesting:

roberthheath.blogspot....
roberthheath.blogspot....

Moody's Doomsday Default Scenario Doesn't Play Out

Nice article, although as I eyeball your first graph, we'd have to see the default rate start declining before concluding whether Moody's estimate of the peak default rate was too pessimistic or simply too early.

Also, I would observe that "... Total return to the High Yield Index (Merrill's Master II) is 40.0% through 8/28, well above last year's -26% debacle...." really just means that total return for the 20 months ending August is 3.6% (.76 x 1.4 = 1.036) consistent with your observations elsewhere that the high-yield vs. investment grade risk premium is hard to find.

Distinction Between Positive and Normative Economics Misses the Point

Nice article.

"Prescriptive" doesn't work because "prescriptive vs. descriptive" is a pretty good way of explaining "normative vs. positive" to students who have no previous exposure to the terms.

"Design" just doesn't feel right either... too vague and too reminiscent of "interior design" or capitalized, "Intelligent Design".

"Engineering" sounds about right to me. Moreover, it provides a context for understanding the limits of a particular science and perhaps a check on regulatory (or financial market) hubris.

Just as a civil engineer designing a bridge will rely on a body of (positive) science to conclude that his structure will function as expected, an engineering economist designing the contours of a cap-and-trade auction does the same. And just as the civil engineer may use an incomplete model (let's say Newtonian physics) which ignores unnecessary detail (like relativity and particle physics), the economist will also do the same (by assuming, inter alia, rationality, complete markets, the law of one price, no collusion, etc.)

A honest "engineering" economist will then consider whether his assumptions can reasonably be expected to hold (and his abstractions can be safely ignored) in the real world before recommending that the system will work as designed. This would be a healthy addition to public policy debates that too often are conducted in sound bites about ends and goals without a realistic discussion of the means by which we get there.

Finally, one of the most interesting subjects in engineering is failure analysis. Watching the Tacoma Narrows bridge fall down (and understanding why) brings the topic alive. The recent financial crisis should provide many opportunities for failure analysis in the sphere of economic (or financial) engineering that will leave your students with equal amounts of understanding and humility at the end of the semester.

Google Is Overpriced: Why Acquire On2 in an All Stock Deal?

You may be reading too much into this.

reinharden's point is a good one. The sellers may prefer Google stock to cash for tax purposes. And if any ONT shareholders want cash, they can easily short the appropriate number of GOOG shares and lock in their value (assuming the deal closes).

Additionally, ONT stock has traded near the $0.60 value of the GOOG deal as recently as mid-May. The management and board may feel that GOOG is a good (or great) strategic partner. It's easier to argue the strategic value of a deal when the consideration is stock vs. cash. If management accepted a cash bid of $0.60 per share, it would be very difficult for them to resist a higher cash bid from an unwelcome suitor.

Case Shiller Index Has First Monthly Increase Since 2006

www2.standardandpoors....


On Jul 28 10:21 AM User 459512 wrote:

> I'm just wondering if the uptick is statistically significant. Does
> anyone know where I can get raw data for the Case Shiller Index?

Larry Summers's Billion-Dollar Harvard Gamble

BTN -

I think you're example actually makes the point that Harvard's interest rate hedge was reasonably sized for its expected borrowing levels. The billion dollar loss represents the present value of many years worth of fixed vs. floating rate interest charges.

The present value of a $100 million interest rate differential (let's say 2% move in rates x $5.0 billion in borrowings) for 15 years discounted at 7% comes to $911 million, roughly equal to the "billion dollar" loss Harvard reportedly lost on the hedge.

Given Harvard's expansion plans, this does not seem unreasonable.


On Jul 24 02:47 PM Between The Numbers wrote:

> Even if it was a reasonable hedge, it magnitude of the trade was
> too much. If Harvard was going to issue $10B of debt (more than the
> endowment of most schools) than interest payments would be ~$500M
> a year at 5%. Even a move in rates to 7% would involve only an extra
> $200M a year in interest, yet he is hedging over $1B. A good hedge
> is one that ends up worthless with the primary investment going up
> more - in this case Harvard would save $200M from rates going from
> 5% to 3%, but lost over $1B. This hedge would be more appropriate
> for $25-$100B in debt, an amount that Harvard would never consider
> doing.

Annals of Rank Hubris, Larry Summers Edition

The Epicurean Dealmaker makes an important observation that misses the mark.

"... I have rarely encountered a corporate client who feels confident enough about both their absolute funding needs and current and impending market conditions to enter into a forward swap starting more than nine months into the future."

Note the modifier "corporate" here. While Harvard University is managed by an organization called "Harvard Corporation" the university should not be confused with a "corporate client."

Harvard has been pursuing the same educational mission since 1636 and probably will be for the next 373 years. Corporations rarely last 100 years, and when they do, it's usually in greatly modified form.

It certainly seems reasonable that with a multi-decade planning horizon and assumptions about the endowment level, future alumni contributions, investment returns and operating expenses, Harvard would have wished to lock down future funding costs well into the future. That would be fairly described as a "hedge" and not "rank speculation."

In the current interest rate environment, the swap position has gone south. But if the intent was to lock in funding costs for a few decades, that's not a terribly important criticism. It's a bit like saying your 401(k) investment in the S&P 500 last year was "rank speculation" because you could've waited and bought it cheaper today.

Harvard's bigger problem would seem to be the other side of the hedge, namely its spending (hence borrowing) plans. With the dramatic reduction in its endowment and probably lowered expectations of both investment income and alumni donations, it may not be able to spend (and borrow) as much as it planned when it entered into the interest-rate swap position. So the (mark-to-market) loss on the swap may not be offset by the funding cost certainty of the future borrowing.

Incidentally, a fair number of astute commentators are worried about the impact our massive deficits may have on future inflation. Homeowners who "hedged out" interest rate risk by locking in fixed-rate mortgages in the late 1960's at sub-5% interest looked pretty smart for the next 30 years. If current deficits lead to sustained inflation, a 20-year interest rate lock from 2006, may look shrewd in a few short years.

True, Harvard got itself into a liquidity bind, forcing it to resort to short-term financing moves in an adverse market in late 2008. And that raises many more interesting questions that I've covered here:

roberthheath.blogspot....

Did Harvard extrapolate bull-market returns too far into the future? Probably. Did Harvard rely on its endowment income to fund too much of its annual operating budget? Absolutely. Did portfolio diversification fail in a global financial crisis as correleations between asset classes increased? Yes indeed. But in this, Harvard looks like any other asset manager (and a great number of US homeowners).

There is a more interesting story here that Nina Munk alludes to in her Vanity fair article. Namely, did Harvard fail (perhaps through inattention) to unwind the swap position in mid- or late-2008 as market conditions deteriorated and long-term borrowing plans might reasonably have been curtailed? As this time period corresponds to management change at the Harvard Management Company, this is an interesting question that Munk asks, but goes unanswered.

But since Larry Summers stepped down as Harvard president in 2006, it's hard to imagine that a fumbled hand-off in 2008 can be blamed on him personally.

Supreme Court's Ruling Deals a Blow to Big Media

"And if this technology isn't a violation of copyright laws, what's next?"

I think the situation is more ambiguous than this, which is undoubtedly why the Supreme Court declined to hear the Cablevision case. You can draw a pretty clear line from the Betamax definition of "fair use" to the "networked PVR" proposed by Cablevision.

en.wikipedia.org/wiki/....

If making a copy of a broadcast show is fair use under copyright law, it shouldn't matter whether the copying device is owned by me and located in my house (like a VCR or DVR) or in my brother-in-law's house, or arguably on a Cablevision server.

That said, I don't think we've heard the last of this issue. When I watch a recorded show on my TiVo, there's no explicit economic transaction involved. If Cablevision explicitly charges for the network DVR feature, it does feel like they're "re-broadcasting" for profit. And even if the cost of the feature is bundled into a monthly subscription, the broadcasters will see no distinction between an explicit or implicit charge.

So at the end of the day, network PVR functionality will likely be offered on terms that (at least some) broadcasters reluctantly find acceptable. This may preclude fast-forwarding through ads, forcing viewers to engage in old-fashioned ad avoidance like going to the bathroom, or to the fridge, or another channel for five minutes.

The cost of storage and video compression technology will continue to plummet, making DVR technology in some form a mass market product. As DVRs are quite a bit easier to operate than VCRs, it seems reasonable that in ten years, DVR may achieve near-universal household penetration (in the U.S.) like VCR did before it.

Forward-looking broadcasters may actually prefer to see networked DVR (with limitations) established as an alternative to forestall standalone DVR penetration (without such limitations). Likewise, cable networks with small audiences may gain meaningful viewership through readily available time-shifting.

I think the Supreme Court is wise to duck this case now. Any fine-tuning of Betamax based on the physical architecture of the network and recording devices involved would likely be obsolete before the law profession could sort out its implications.

Moreover, if over-the-air broadcasters -- who currently enjoy "must carry" privileges on cable systems -- move towards negotiated carriage fees as CBS has suggested it might, one can reasonably expect that network PVR rights will be part of that discussion.

During this period of rapid technological and regulatory upheaval, the Supremes are wise to let the market sort this out.

The Unraveling of Newspaper Economics

259326 -

Thanks for your kind comments.

I wrestled with the issue of how to present cause-and-effect regarding bundling and monopoly.

It's certainly easier for bundling to work in the presence of a distribution monopoly and in the extreme case, bundling won't work in a purely competitive and transparent market in which the bundled goods are priced separately and more cheaply. But bundling doesn't require a monopoly to work as a pricing strategy.

As I recall it, bundling can be effective if consumers have heterogeneous demands (you prefer the business news, I like the comics and weekend section) and the firm cannot price discriminate.

In the newspaper industry, it's certainly true that newspapers bundled all manner of content back in the day when most cities had at least two major papers or more.

My diagram suggests a one-way causal relationship, which is a limit of a two-dimensional diagram and perhaps my creativity. In reality, I think all these factors reinforced each other in a "complex" and self-reinforcing relationship.

And yes, "surplus economics" is "profit". I chose the former term to emphasize my belief that even if newspapers could instantaneously transition their businesses to the web, eliminate their newsprint and physical distribution costs and successfully deploy micro-payment and subscription models, it still might not provide the same profit as their old business model because they can no longer expect the economic benefits of bundling third-party content that's readily available (often for free) elsewhere on the web.

Thanks again.