Tom Armistead's  Instablog

Many pundits have opined that the American consumer is on the ropes, so far in debt that he will be unable to contribute to the economic recovery. Because consumer spending has recently been about 70% of GDP, this has led many to declare that the recovery cannot proceed until a lengthy de-leveraging process has been completed. Factual information is available to assist in quantifying this concern: here is a link to the Federal Reserve's Household Debt Service and Financial Obligations Ratios. After downloading and studying this information, a plausible case can be made for the possibility that this process can be completed in a one or two year time frame, after which a strong and sustainable economic recovery would be possible.

The household debt service ratio - DSR - is an estimate of the ratio of debt payments to disposable personal income. Debt payments consist of the estimated required payments on outstanding mortgage and consumer debt. The Fed provides information going back to 1980, so it gives us a look at a number of troughs/recessions which preceded strong recoveries.



How much debt is too much? - The average ratio from 1980 to the present is 12.06%, with a max of 13.90% in 3Q 2007 and a min of 10.60% in 1Q 1983. My impression, after eyeballing the data, is that a DSR of less than 11% is optimal as a prelude to a strong economic growth spurt. Consumers are not over-extended and can safely handle more debt and drive the economy. This condition was present in the early 80's and again in the early 90's, when strong recoveries resulted. In the early 2000's and subsequent, DSR was not meaningfully reduced, which may be attributed to the extremely lenient monetary policies in place at the time. Very possibly this de-leveraging was not avoided but only deferred, and is taking place now.

DSR as of 2Q 2009, the last data point available, stood at 13.11%, and has been declining since the high of 13.90 % in 3Q 2007. Is it possible for DSR to be brought down as far as 11%, and if so, how long would that take? How about 12%, a little below average?

Making it better - to improve the ratio, either disposable personal income must increase or debt service must decrease, or some combination of the two. DSR was low in the early 80's because interest rates had been pushed so high nobody borrowed if they could help it. Also, the horrible inflation at the time had one benefit in that it allowed the repayment of debt with depreciated dollars. So that laid the groundwork for a strong recovery, in the economy and in the stock market. However, if we were to hold debt service steady and rely on increases of disposable personal income, 3% a year would require 6 years to get DSR down to 11%.

Debt service can be reduced by paying off debt, by restructuring it, or by writing it off. Banks have been taking huge write-offs on mortgages and credit cards, but more will be required before this is over. There has been some restructuring of mortgages, and some refinancing at more favorable terms. It is unlikely that banks can or will write-off or restructure enough debt to drop the DSR as far as would be desirable. Consumers are reducing debt, but if your credit cards are maxed out it takes a long time to dig out of that hole, particularly when every late payment generates another fee and another rate increase.

Actually increasing disposable personal income by increasing employment, rather than by igniting inflation, would be the ideal solution, but appears unlikely in the near future.

Projecting from the current rate of progress – during the first 2 quarters of 2009, DSR was reduced by .45 %, or .23% per quarter. At that clip, to get from 13.11% to 11% would require roughly 2 years, assuming things go on pretty much the same as they have been for the first half of the year. To get down to an average DSR of 12.06% would be about a year, maybe somewhat more. The third quarter information will not be available for another month or so, but it will be very interesting to see if the trend of reductions continues.

Investment implications – if consumer spending power requires another year (or two) of de-leveraging/recuperaton, it is unrealistic for investors to look for a strong economic recovery until that process is complete. The Fed's data linked above provides an objective way to monitor developments. Possibly easy money policies will again paper over the DSR situation, leading to a premature and unsustainable recovery. Or the process of de-levering, once started, may be self-sustaining, setting the stage for a better recovery.

Thoughts on public policy – enlightened public policy would attempt to bring DSR down to levels that have been associated with stable growth in the past and keep it there. That would imply diligent prudential regulation of the banking industry to ensure that they do not lend under terms that impose excessive debt service obligations on borrowers. Alan Greenspan would have done well to consider DSR before continuing his easy money policy beyond its proper duration. 

Disclosure -  No positions mentioned, the author is net long stocks 

Eric Dinallo was on Bloomberg recently, talking about CDS and other regulatory issues.  This clip www.youtube.com/watch is worth watching: he is the only politician or regulator who understands the dangers posed by CDS and inadequate regulation and is willing to talk about realistic solutions.

Dinallo, as former NY State Superintendent of Insurance has first hand knowledge of the issues involved, due to the problems he dealt with on AIG, MBIA, Ambac and other financial guarantors.  He and NY Governor Paterson attempted to bring the CDS issue to a head by offering to regulate them as insurance, if the Federal government did not step in and do something.  At the time, it seemed that Congress would be doing something, but more than a year later CDS regulation is still being debated and in the process whatever we get is going to be too watered down to resolve the issues.  Dinallo makes some excellent points: 

CDS should be regulated - as either the insurance products they are or the gambling products they are.  Approximately 20% of CDS are supported by an insurable interest: they should be regulated as insurance with requirements for adequate capital.  The other 80% of CDS are gambling contracts and should be regulated as such.  

These views are in sharp distinction from those we see in Congress, where there is considerable weight given to the input from Goldman Sachs (GS), Bank of America (BAC), Morgan Stanley (MS), JP Morgan (JPM) and CItigroup (C).  All of these "banks" make good money trading and steadfastly resist any efforts to regulate CDS in such a way as to prevent a recurrence of the debacle that required their rescue on the backs of over-burdened taxpayers. 

CFMA, Graham-Leach-Bliley, Glass-Steagall - CFMA was the legislation that exempted CDS from regulation by either the CFTC or the SEC.  Glass-Steagall enforced the separation of banks from other financial activities such as insurance, and Graham-Leach Bliley removed these important protections from the regulatory arsenal.  

Dinallo's common sense analysis: depositary institutions must not be permitted to indulge in leveraged financial bets.  That would include insurance companies such as AIG. Basically we need to repeal CFMA and Graham-Leach-Bliley, restoring the legal basis for adequate regulation of financial firms.   

Compensation for TARP recipients - after noting that Wall St. seems to have trouble grasping the implications of the compensation issue, Dinallo adds that it's not just about the top executives.  There are many operating employees (traders) making the big bucks.  Obviously their compensation schemes motivate extreme risk taking, inappropriate with taxpayer support.  

But he sees the solution as being improved regulation, not setting arbitrary limits on compensation.  If CDS are limited to their proper function as insurance, issued by institutions that have adequate capital, the problem of excessive compensation will go away.  If banks and insurance companies that are entrusted with people's savings and financial security are prohibited from gambling, the high rollers will go elsewhere, where they can't endanger the system, and their compensation will be paid by their gambling adersaries rather than taxpayers.  

NY Attorney General - DInallo is a candidate for this position.  Noting that Elliot Spitzer was able to act as the "Sheriff of Wall Street" while serving as NY AG, and that DInallo makes sense on a lot of regulatory issues, there is potential for some reduction in Wild West capitalism.  

Disclosure -  No position in the firms mentioned.  
 

Sigma Designs (SIGM) recently announced the acquisition of CopperGate for 160 million, to be paid by 92 million cash plus the issuance of an estimated 4 million shares. Boards of both companies have approved the deal. As has been happening lately, shares of the acquiring company traded down. Sigma closed Tuesday at 13.46, down about 6% on the day. After reviewing the slides that accompany the conference call, I regard the acquisition as a favorable development, based on a good strategic fit and the ability to fund it without incurring debt.

I wrote SIGM up favorably on 9/14 at 15.67, with my case resting on the possibility of renewed growth by expanding its IPTV expertise from telcom into the cable area, together with margin of security due to the excess cash on the balance sheet. A number of commenters questioned the pick, and the discussion got to be about whether management would dissipate the cash trove by high price buybacks or by overpaying for a series of ill-conceived acquisitions. A recent article on SA suggested SIGM would be a good target for an activist shakedown, adding a special dividend to the mix of possibilities. The acquisition of CopperGate answers these questions.

The Deal – from the press release:

Sigma Designs(R), Inc. (NASDAQ:SIGM) ("Sigma") and CopperGate Communications Ltd. ("CopperGate") today announced that the companies have entered into a definitive agreement for Sigma to acquire CopperGate in a cash and stock transaction with an agreed value of $160 million, net of CopperGate's cash at the closing of the transaction.

The estimated amount of cash to be paid by Sigma on the closing date is approximately $92 million, plus the amount of cash and cash equivalents estimated to be held by CopperGate at the closing, net of CopperGate transaction expenses and debt outstanding at the closing. In addition, Sigma will issue shares of its common stock to CopperGate shareholders estimated at the time of signing to equal approximately 4.0 million shares.

The combination of Sigma and CopperGate is also expected to yield several potential synergies including synergies from leveraging manufacturing know-how and combined wafer sourcing, further SoC integration and combined research and development.

Side by Side – from the presentation slides:

	Sigma Design	CopperGate
	TTM 8/1/09	TTM 6/30/09
Revenue	$197 million	$69 million
Gross Profit	$91 million	$39 million
Gross Margin	46%	56%
Income from Operations	$17 million	$10 million
Income from Operations - Margin	9%	15%
Net Cash	$229 million	$24

Doing the math, the combined entity has 266 million of sales and proforma TTM earnings which I estimate at.88 per share. Most of SIGM's excess cash has been spent on the deal, so valuation at this point is very much dependent on estimated future EPS growth. On the topic of synergies, R&D was 23% of SIGM's revenue during their most recent quarter. If Coppergate's expenses were comparable, any overlap of R&D that can be eliminated would result in substantial expense savings. Because R&D is closely related to the sales process and has an 18 month cycle, making changes in an orderly fashion would require about 2 years to implement. CopperGate is approximately 1/3 the size of Sigma, so the acquisition is large enough to be meaningful without being so big as to pose difficult integration problems.

According to the presentation, worldwide broadband households will have a CAGR of 11% from 2009 to 2013. CAGR for telco IPTV cumulative subscribers will be 36%, for worldwide home networks 15%, and for worldwide network enabled video devices 65%. All of these estimates are double digit, some more impressive than others.

Assuming SIGM participates, double digit growth is implied.

Valuation - Using the .88 EPS estimate, and applying a P/E of 23 for a double digit growth situation, 20 is reasonable if and when SIGM demonstrates the ability to participate in the industry growth projected in their presentation. As discussed in the original writeup, rapid revenue increase in a fabless operation with high gross margins can generate increased EPS very rapidly, which in turn generates P/E expansion, with gratifying results.

SIGM has traded in a very broad range of price/sales, from .9 to 11.2 over the past 5 years. Ken Fisher did quite a bit of work on P/S for his 1984 book “Super Stocks,” noting at the time that higher price/sales ratios are common in growth stocks with lower market caps. From 2004 to 2008 SIGM traded at a price/sales over 7 at some point during every year. The combined entity will have TTM sales of about 9 per share, multiplying that by a historical midpoint price/sales ratio of 3.5 suggests a price of 31. Again, this depends on the resumption of robust revenue growth.

These recent ratios are consistent with Fisher's advice from the 80's: he suggested buying companies of this type at P/S ratios of less than 1.5, .75 would be better, and selling them somewhere between 3 and 6. At a recent price of 13.05, the proforma combination trades at a P/S of 1.4, falling within Fisher's guidelines, although cheaper would be better.

Noting that this is an acquisition involving companies that have been profitable in the difficult recessionary environment, with no long term debt, downside risk is moderate while the upside potential is large but indefinite.

Any acquisition raises the question of possible overpayment: however, my view is that if the acquiring entity is able to consummate the deal without incurring debt, and if the strategic fit is good, then even a generous price will prove to be money well spent. Shareholders who contributed the funds which have been sitting idle on the balance sheet intended that the capital would be deployed in the IPTV/SoC business. After initially buying back some shares, then doing a small but strategically questionable acquisition, management has now put the money to its intended use, a positive development.

Some P/S math: during its fiscal 2008 year, SIGM issued 4.6 million shares at 43 per share, which valued the company at 7 X sales. SIGM has used these same funds to acquire CopperGate at 2.3 X sales.

PRR – Price to Research Ratio - Fisher had another rule of thumb, applicable to these cases where R&D is a substantial part of the mix. “The Price Research Ratio (PRR) is the market value of the company divided by the corporate research expenses for the last 12 months.” His rules: do not pay more than a PRR of 15; preferably these type companies should be bought somewhere between 5 and 10 on this metric.

Sigma's market cap is 349 million, divide by 2008 R&D of 43 million and you get a PRR of 8.1, within the suggested range. Guessing that CopperGate has R&D of 10 to 15 million, that would imply a PRR of 6 to 6.5 for the combined entity, right about where Fisher suggests is the sweet spot.

Short interest – as noted in the previous writeup, this company is heavily shorted. With the number of shares to be issued to pay for the deal an open item, short-sellers might try to force the price down ahead of the closing, thereby forcing the company to further dilute its shareholders. On the other hand, management has enough extra cash to increase shareholder value by buying some shares back if the situation gets out of hand.

Strategy – the October 15 calls I sold over my position will expire worthless. Noting the shares are declining on the acquisition, which I regard as a positive, I have been enlarging my position. I plan to monitor earnings and conference calls, looking for confirmation that Sigma will be able to penetrate the cable market for IPTV and triple play.

Disclosure – long SIGM