The next debacle - taxes after the mortgage mess

60 Minutes had an interesting piece (below) on the low-end of the mortgage crisis - (1) borrowers bought houses with no money down (and often with cash back), (2) lenders, brokers and everyone else pushed them to do it because commissions were based on the close of the transaction, and now that home prices are falling (down 21% YoY in the Bay Area for the bottom third of home values), and (3) buyers, unable to sell or refinance since they now owe more than their home is worth, are defaulting (with >$1B of homes in foreclosure just in Stockton, CA). All the losses get triggered on the banks' books because the debt gets downgraded, so the value of the CDOs (collateralized debt obligations = pools of mortgages) they carry plummets and they take those losses now.

S&P estimates more than $265B of total losses from the CDO write downs. That's a huge mess.

The next mess is how all of that wreaks havoc on the tax system:

1. People who short sell (sell for less than their mortgage(s)) or have their homes foreclosed owe money to the IRS (and their state) in two ways - (1) they may have gain on their house. Even if they get foreclosed and they owe more than the house is worth (because of home equity lines or refinancing), they could still be "selling" for more than they paid. That's capital gain, but on a primary residence most people (in these circumstances) can exclude it for tax purposes; (2) if the mortgage company doesn't collect all of the loan, the part they "forgive" counts as income. So a $300K mortgage on a house that's short sold for $200K, generates $100K of taxable income for someone who's in the $50K range on annual salary (even worse, it puts them in a higher tax bracket) - so they might owe $35K of tax on the loan forgivenes, even though they have no money to pay. The tax liability gets forgiven if they go bankrupt or insolvent. (So maybe the "best" plan is to run up tons of other debt and make sure the tax liability pushes you into insolvency!) In most cases, the IRS will never be able to collect, but they're going to try and it's going to give a lot of people heartburn. It's a good time to be a low-end tax attorney. This is a major personal tax problem for the people in the 60 minutes segment who can pay but just think paying doesn't make sense - the IRS will collect in that case. (Maybe they prefer to give their money to the government than to the bank.) It's not much of a federal tax problem by not collecting, since the government never anticipated that money as income (except for point 2 below). Of course, if the government just bails people out of it by waiving the cancellation of indebtedness provisions, most of us will never work for salary again - we'll just have our companies loan us money and forgive it later. It's definitely income (the 60 Minutes segment describes people who have spent proceeds from the forgived loan, just like they would spend any other income), so it seems fair to tax it.

2. The Federal government loses on the tax writeoffs that the banks with $265B in losses are taking. At the corporate tax rate, that's about $70B in lost taxes to the feds (some of it's foreign held, etc, but putting tha aside...). This is why the forgiveness in point 1 gets taxed - because the act of forgiveness is a writeoff for the person doing the forgiving. They could just not allow it to be a writeoff anymore so that the banks pay, but then you can bet the banks will then work harder to collect on short sales. Credit would also get a lot tighter. There's no easy fix, so here's $70B (minus whatever they collect in tax revenue from individuals in point 1) lost to the federal government.

3. The states are the big losers. They lose a lot of property tax revenue - not just 1% * $265B per year because most home values are falling - so people can get their homes re-assessed whatever their financial status. It doesn't matter whether they move, get foreclosed, stay put, whatever. This is an annual shortfall, and it may be quite awhile before home values get back to where they were - so states will have less money for quite awhile. Of course, when they budget, they don't plan for that, so there will be a bunch of state budget crises for the next several years because of the mortgage mess. Cleveland and Baltimore are even suing over the lost tax revenue. (They're not giving back the excess tax revenue they earned from builder taxes, payroll taxes, or proerty taxes while the "defective loans" were being made, so I'm not sure their suit is exactly symmetric, but they have interesting arguments.)

Although it's painful, and unfair if there was predatory lending involved, I think it does "make sense" that they have to pay their mortgage even though their house declined in value. It is what they agreed to do (if they can't then bankruptcy is an alternative, but they say they can.)

Banks are going to raise borrowing costs for people who walk away (and probably for everyone else), so we'll all be forced to live within our means. Not necessarily a bad thing, but it sure doesn't bode well for consumer spending! (HELOCs are already getting shut down.) The tax issues are going to clamp down state spending, and the federal government is already in deficits.

From basic macro, GDP = consumption + investments + government spending + net exports.

GDP and government spending are headed down. Time to find something recession proof to do.

Technorati: mortgages, subprime, recession, taxes

Search is NOT recession proof

I read a MediaPost summary of a JP Morgan hosted call with DidIt where the question of whether search is recession proof came up. It's not. The common argument (and the point DidIt made) of "the ROI is there and our clients are self funded above certain ROI levels" misses the boat. Henry Blodget points out that query volume declines during a recession, so even if the ROI is constant in a campaign, the search spend declines (since there are fewer searches to spend it on). I disagree slightly - with fewer searches, the initial impact will be lower ROI since the same number of bids happen across a reduced supply of searches will make bid prices go up. This will eventually also lead to lower search spend as companies cut budgets to keep ROI constant (getting to the same place that Blodget points out.)

There are plenty of other reasons search isn't recession proof: (1) Conversion rates will go down, as more people window shop. This will directly impact ROI, and again, lead to lower search spend. (2) Not every search advertiser is ROI-savvy, but in a recession, more of them will be forced to scrutinize ROI. (3) Plenty of small businesses (the long tail of advertisers) will fail in a recession and be unable to spend on search. (4) For websites that make money on leads or advertising, the value of those leads and advertising pages will decline (as their customers cut back because leads and ads aren't converting), so their ROI will decline unless they cut spend (since their top line will go down).

Search is definitely NOT recession proof. Maybe Google the stock (GOOG) is recession proof because (a) they can continue to gain share faster than spending declines, (b) they may have other businesses that are growing faster than spend will decline, or (c) spend is still pouring on to the web faster than the decline in the spend that's already there. (Although even these would still mean slower growth for Google during a recession than they'd otherwise have.) I have no idea whether the stock is recession proof, but the search business certainly isn't. If I were still in it, I'd be planning accordingly. Not hoping that it's unlike other businesses - that feels eerily reminiscent of the "new economy" chatter from 1999-2000.

Technorati: Google, search, recession

Cheap mobile office setup - updated

In September, I set up a mobile office for $1K. Emoze, the software I used to simulate push email on my WM6 device (a Tmobile Dash), was ok, but it wasn't great. It had to be running on both my laptop and my phone (neither of which has a ton of memory), if my laptop hibernated it would take down my email link, and when I moved items around on my calendar, Emoze didn't always keep up. I only had to go to the wrong place at the wrong time for a meeting once to start looking around for an updated setup.

At first, I shifted to Hotmail. Hotmail is really well integrated with WM6 - on the surface. It pushes email, syncs the device to the web, is easy to set up. Unfortunately, it has a few quirks: (1) When you send an email from the WM6 device, no "to" field shows up on the Hotmail website. So the email goes through fine, but you can't easily find it in Hotmail by sorting later; (2) When you send an email from WM6, it doesn't adhere to the "from" field you set up at Hotmail.com. I didn't want people to use my hotmail address, since I wasn't sure I'd stay, so I set up to show the "from" line using my alumni account. That worked fine from Hotmail, but strangely, not from the mobile phone (which is supposed to be synced with Hotmail). I could have lived with those, I think, but the TrueSwitch service, provided by a third party through a deal with Hotmail, was absolutely terrible. TrueSwitch was supposed to transfer my Gmail in a nice organized way to Hotmail. Instead, it forwarded 50,000 messages to my Hotmail inbox, didn't organize any of them, ad then just stopped (so not even all of my messages made it over). After a few emails to tech support (where they promised a 36 hour resolution), I got a response saying they didn't know what happened and would have to send it to engineering. Nice customer service. Nice partner you've chosen, Microsoft!

I went back to Gmail and set it up for IMAP access that is native on the WM6 phone. It's not quite push, but it checks frequently enough that it might as well be (that hurts my battery life a bit, but it's livable). Unfortunately, I also have to use the Gmail Java WM6 application to see some emails, because there's a known bug in Gmail IMAP that prevents HTML messages from showing on WM6. At least I can see the header on something like a push basis and switch to the Java app if necessary. The integration of Gmail IMAP and Outlook isn't great - items deleted from my phone go to a deleted items folder, but not items deleted from Outlook, for instance - but at least I have the Dash, the web, and the laptop all synced on email. When Google fixes the HTML email problem, this should be a pretty solid setup, without the extra overhead of Emoze. Without Emoze, I don't have my calendar synced to Outlook wirelessly, but I'm working on that - no sync is better than faulty sync.

During this switchover, I was struck by Microsoft's squandered opportunity. Here I am ready to switch from using Gmail, which is not meeting my needs, to Hotmail. I use a Windows PC, I'm addicted to Outlook, and I have a Windows Mobile cell phone. If I switch, I'm locked in to all Microsoft all the time. But their integration is so clunky - need the Outlook Connector to get Outlook connected to Hotmail, inconsistent use of the "from" and "to" fields across the three platforms, no free calendar sync, and the partnership with Trueswitch that made my Hotmail account unusable - that I had to go back to using Gmail, even though it has some quirks too. Google will have its own devices soon and will likely make the integrations smoother, so if MSFT doesn't take advantage of opportunities to capture users like this now, they may be too late.

Technorati: cheap mobile office, Trueswitch, Gmail, Hotmail, Microsoft, Google

Audi introduces the PRUGO5

(DETROIT) Today at the North American International Auto Show, Audi announced immediate availability of the PRUGO5. The PRUGO5 beats the comparably equipped BMW M5 in 0-60 and quarter mile performance, improves on the industry-leading fuel economy of the Toyota Prius, and comes fully loaded for about half the price of an old Yugo. The car won't work for everyone - towing capacity is limited and it only seats 5 adults comfortably - but nearly all customers polled are excited about the feature set. Audi expects to sell 6M units in the first year, breaking every previous sales record, and quintupling their market share. ...

The PRUGO5 would be highly differentiated in the marketplace. It's tough to imagine a salesperson struggling to sell it to anyone. It's also equally tough to build, and even tougher to build cheaply enough that it sells for a profit. Unfortunately, many product specs, guided by salespeople who want to make their jobs easier and their commissions larger, end up looking a lot like the PRUGO5. Rather than make tough decisions, product management just asks for the product to do everything. This leads quickly to a major breakdown between product and engineering, with product (and sales) blaming engineering for being unable to build stuff and engineering blaming product (and sales) for pie-in-the-sky requirements.

Product management is the central nervous system of a (product-driven) company. Product managers need to understand not just the market requirements but also the tradeoffs in those requirements, the willingness to pay of customers in each segment, and the complexities inherent in building solutions for each segment. Getting those answers is tough - even impossible in many cases of brand new products where there is no market data - and even when answered, salespeople will constantly push for feature creep anyway. But the difficulty of getting those answers is also why product management can be so rewarding and can have so much influence over the direction of the company.

There are plenty of tools to understand market tradeoffs and willingness to pay across segments - conjoint analysis, regressions, observations of changed usage across small variations of the product, etc. Even trial and error work fine - start with a hypothesis, figure out a fast way to test it (e.g. put screenshots and sample prices in front of customers), and then keep going.

In finding these tradeoffs, I've seen product management struggle in two major ways: (1) Asking customers "do you want this?", and (2) Asking customers "what do you want"? The problem with (1) is that there's no cost for the customer to say yes - who doesn't want the PRUGO5? Much better to ask a bunch of "would you prefer (a) or (b)?" questions (a conjoint) with lots of followup about how they'd use each potential product. The problem with (2) is that it allows the customer to stay bounded by their current experience - so you can make gradual, incremental product improvements but can't ever have a discontinuous innovation. If you can suggest something discontinuous, maybe even show how the customer would interact with it, and then ask followup questions, the data you get back on new products is much more useful.

Understanding the market requirement tradeoffs is just the first step for product people - reducing all of that to a profitable line of business, also a responsibility of a good product manager, involves several more steps (and more blog posts). Without having the tradeoffs understood, though, the rest of product management is mostly irrelevant.

(Thanks to Shawn Gradek, one of the nicest guys and best salespeople I know, for the name of the car.)

Technorati: product management, entrepreneurship

Many lessons (partially) learned

I've had a crazy run during the last 18 months - from CEO of a venture-backed startup just beginning the sales process, to completely unempowered GM of a group after the company got acquired, to CEO of another venture backed company in a completely different field, with a two month stint as an interim turnaround executive at the company that acquired mine thrown in for good measure. The jumble of experiences has shaped (warped) my perspective a bit, and writing down my thoughts as my perspective changes helps me reflect and learn. I'm sure that after the next group of intense experiences, many of my "lessons" will morph again, but for now, this is where I stand. As I reflect on various topics, I'll add hyperlinks to the posts below.

I'll start with Product Management (Audi introduces the PRUGO5!)

[Added 2/5/08] - Management's role in building a good culture.

Technorati: entrepreneurship, venture capital

RevCube is going under - my prediction from LinkedIn

RevCube Media is going under - maybe not immediately, but sometime this year. RevCube works to optimize many pieces of online advertising across multiple channels - a similar goal to ClickShift (now WebTrends Dynamic Search), which I started at about the same time. I was introduced to the founder of Revcube (started as AdaptAds, but forced to change its name by Adapt SEM, another company entering the same space at about the same time) in mid-2005, just as I was starting ClickShift. In a "Silicon Valley is a very small world" story - the co-founder of my current company, who was in my wedding, made the introduction at his wedding.

While updating my profile on LinkedIn and seeing who my friends had recommended, I noticed a flurry of "recommendations" for a RevCube exec in the last few weeks. Clearly, the request for references was sent while still at the company. That means a departure is imminent. But since the references were from other Revcube people, leaving is no secret - so execs at RevCube know that other execs are looking for work and are proactively helping them find it. The company must be in trouble. If it weren't, it would be difficult for an exec to leave with the proactive help of the CEO. The investors aren't known for being "entrepreneur friendly", so I don't think they'll provide much support in the hard times. The company will go under this year. If I were thinking about becoming a customer, I'd want to see some financial statements at the very least.

The best competitive situation is when several of the startups in the space thrive - means you picked a good market - so I don't enjoy watching RevCube go. But I'm pretty sure they're on the way out.

Technorati Tags:revcube, linkedin

Worst venture firm name ever

I noticed this in PE Week Wire today:

"Confirma Inc., a Bellevue, Wash.-based provider of computer-aided detection for MRI , has raised $17.5 million in Series C funding. Telegraph Hill Partners led the deal, and was joined by Fluke Venture Partners and return backers Northwest Venture Associates, Prism Ventureworks and Versant Ventures. The company previously had raised around $18 million since 1998. "

I realize it's a family investment fund and that it's had good success, but "Fluke"? It seems like the family would have been more willing to go for an uplifting name, like Ridgelift, Redpoint, Foundation, Sequoia. Anything would suggest more success and stability than "fluke".

WebTrends settles with NetRatings!

Webtrends announced on December 31st that they settled their patent litigation with Netratings. Omniture (OMTR), Visual Sciences (VSCN), and Coremetrics have already settled. Congratulations to James McDermott (the WebTrends GC) who has been working hard on this for a long time!

Netratings is a bit of a patent troll, and the patent system could use some reform for both prosecution and litigation, but this seems like a nice outcome for everybody.

Technorati Tags: Webtrends, Netratings, patents

It has been a crazy two months!

Just when I had resumed blogging again, I stopped to return to WebTrends on an interim basis to help with the management transition. It was a phenomenal experience - the people there are great, everyone was supportive, the company has a huge amount of untapped potential, and seeing the repercussions of previous bad management decisions is about the best way to learn how to manage effectively. I hope the people at WebTrends got even a tenth as much out of my return as I did. I think 2008 is going to be a good year for the orange team. (But for my investing rules, I'd even contemplating going short OMTR - with an effective management team now in place, WebTrends will be a formidable competitor).

I hope to blog soon about the various lessons I learned in my short stint as an interim executive. Meanwhile, my new insurance company is up and running - the venture financing closes Tuesday, and we've had a positive initial meeting with the regulators who need to be involved. I had hoped to take it slightly easy for a few weeks, but it looks like it might move a bit faster than planned.