Twenty Third Floor
Technical Business Advantage

Moneyweb has an article describing the failure of the South African insurance industry to provide insurance to the wider population, including lower income markets such as the banking sector has done.

There are some interesting points to discuss here, and I’m certainly not saying the industry could not do more. However, there are some fairly fundamental social, pyschological and technical reasons that need to be overcome first. I’ve repeated some comments on the article below. I don’t claim this to be an exhaustive list, but I suggest that it lists some more likely suspects for the causes of imperfect penetration of the insurance market.

Sorry Felicity, but this article doesn’t even get into the details and shows a lack of understanding of the drivers of the need for insurance, and the perceived need for insurance.

Life Insurance
Comparing insurance to banking is disingenuous. Transactional banking makes your life easier, now. Basic savings account can work towards short-term goals. Life insurance will always seem less pressing.

1. Savings products will not work for lower income policyholders through an insurance policy because of the assumed average tax rate of 30%. It is a good deal for welathy investors in high marginal tax brackets, but awful for poor people. This is a function of the tax system not the insurers.
2. Life insurance requires payment of a premium now for a possible future benefit to dependents. There is no way this will ever be a priority need. This is human nature. Even if policies are sold, they will be lapsed very quickly and “better” uses are found for the premiums.
3. Lower income market segments typically have greater reliance on extended family for support. Thus, the need is lower for insurance. This is typical of developing countries, and declines as wealth and education increase (along with smaller families, later first children and less support from the extended family).

4. Funeral insurance may be sold through non-traditional outlets, but it is still exactly life insurance. Just that here the need is better appreciated and understood. Therefore it sells. Or do you want insurers to sell products for which there isn’t a need. (hey, easy on the comments that they already do… I don’t think insurers are angels!)
5. Credit life is required to protect the lenders from the death of the borrower. Again, there is a clear need and this form of insurance is quite widespread. Incidentally, funeral insurance and credit life are, unfortunately, typically quite profitable business lines. This might be a better line of attack against the insurance industry.
6. Insurance in South Africa has remarkably high penetration as measured by insurance premiums as a percentage of GDP and compared to other countries. This shows the succes of the industry, and also explains the limited growth prospects. Life insurance is typically less prevalent than short-term insurance in developing economies – if the problem isn’t restricted to South Africa maybe we should look for broader reasons?

7. I know several insurers who are targeting lower income markets with mixed success. The typical complaint against insurers is that they are overly profit-seeking. If (if!) this is true, then one can’t also complain that they aren’t following up on profitable opportunities? Again, maybe the reason is broader than you’ve implied.

Short-term insurance
Several other commentors have already described valid reasons for why short-term insurance take-up is lower than might be hoped. In many countries, third party liability cover is a legal requirement to drive a vehicle, and with good reason. This is the case in Lebanon, another country where I understand a bit about the insurance industry.

Losses on equity portfolios for our short-term insurers don’t really translate to a requirement to provide insurance to new markets. Maybe it suggests a requirement for less reliance on equity bull markets for performance in good years.

Short-term insurance in South Africa would be considered competitive by most standards. If there were large, profitable, untapped markets out there (with sufficient volumes, limited fraud and low enough claims frequencies and severities to make the premiums affordable to the target market) I expect they would be aggressively pursued. The thing about third party liability cover is that it isn’t greatly a function of the value of your vehicle. That makes it relatively expensive compared with the value of a car typical of a lower income target market. Being insured against someone else’s costs, when you would have no way to pay them otherwise and therefore it would be pointless to be sued, doesn’t sound like a very likely expenditure item.

The expenses of adminstering a policy are also not related to the size of policy or the value of insured property. One can argue whether current efficiency levels are right, but that is a separate argument (and one likely to suggest job cuts…).

The propotion of South Africans with short-term insurance should also be compared against those with sufficient assets to make it sensible. Direct comparisons against the populatio as a whole are close to meaninless.

Ok, I think I have done more than enough rambling and ranting. However, let me conclude with one observation on a quote from the article:

“And the plain fact is that local insurers have done way too little to develop products that offer value for the vast majority of South Africans. This is self-evident; precious few South Africans use insurance products.”

Just saying something is a fact doesn’t make it a fact. And please don’t abuse “self-evident”. Just because one item could be a cause of something does not make it the cause, or the only cause, or the primary cause. Especially not when you have just laid out a few of the reasons I also covered as to why insurance is a hard sell.

I wonder whether I should have mentioned the bad debts on house and car loans that are stacking up based (only partially!!) on pressure to lend to households with little wealth for large deposits and strained financials?

From FT.com

Official data showed the UK economy contracted 1.5 per cent in the last three months of 2008, the biggest quarterly slump in 28 years. The figures confirmed that Britain had entered its first recession since 1991. The figures, a preliminary official estimate of output, showed a contraction in services, manufacturing and construction.

But that 1.5% is much worse than it might sound. We’re used to hearing Seasonally Adjusted Annual Rates (SAAR). But this is a 1.5% decline in a 3 month period. In the US or South Africa, where GDP is reported on an SAAR basis, this would be equivalent to a 6% decline.

Sounds more serious now, doesn’t it?

What do the names Oneway, Bull, Kansvatter, Inthemoney, The Bull! and Millionaire in the making have in common? They’re all optimistic, aggressive names used in the Make A Million competition. They’re also, as of right now, all in the bottom ten out of nearly 300 entrants and have all lost everything but a few rand from the R10,000 they started with on their chance to Make A Million speculating with Single Stock Futures. None can even take what’s left of their R10,000 and watch a Movie (except maybe at a Sterkinekor Classic, on Tuesdays, in PE).

These aren’t the unlucky few. As I warned in a previous post on how the Make A Million competition is a bad idea and encourages wild speculation and ill-advised risk-taking, the performances of the best and worst entrants is dramatically different after just two months, with rather more entrants looking at poor returns.

The highest return (so far as of about the time this blog is posted) based on the live leaderboard is a massive, impressive return of 735% in two months. That’s over a 30 million percent return on an annualised basis. That is also where the good news ends.

Some more stats:

• Highest return 735%
• Average return -27%
• Median return -53%
• Worst return -107% (yes, someone lost all their fund and more thanks to the geared danger of SSFs)
• Approximate percentage of entrants with negative returns 82%

It’s abundantly clear that entrants have been massive losers in just a short space of time.  The average return translates to an annual return of -85%. Of course, the position would have looked much better had the markets boomed during the period, and it’s still possible there could be a huge rally from now until the competition ends. It’s just that I wouldn’t give any of my money to Bull! or Spitfire or Druggies or even Fantastic.

Not in a million years.

Bodie, Kane and Marcus wrote a wonderful textbook used in finance in universities around the world. I remember reading one of the “boxes” during a lecture at UCT many years ago. It described an ingenious money-making strategy:

Start 16 apparently independent investment advisory newsletters. In the first year, 8 of them forecast gold (or oil, or an individual share) to go up, and 8 of them predict the same security or indicator to fall. At the end of the first year, half of the 16 will have given the correct prediction. The incorrect 8 publications are discontinued.

Out of the 8 remaining, 4 predict a certain (maybe the same) security or indicator to go up, and the other 4 down. At the end of the second year 4 of those will have been right. By this method, at the end of 4 years (or 5 years if we start with 32 publications and so on) we will have a single publication with a perfect track record.

The money-making comes from then selling the next year’s publication at a huge price based on it’s apparent track record.

It’s pure genius, except that it’s illegal (very specifically in the US and clearly fraudulent in general terms in any country I can think of). The strategy may even have a name.

I blogged previous about how the Make a Million competition has, IMHO, dubious merits. However, there is a new twist.

Possibly in response to my concerns around the risk-taking nature of the competition (ok, I doubt this was there motivation, but it’s a nice thought) they describe an alternative strategy that can be implemented to potentially win the competition without taking direct market risk.

The basic strategy involves creating two separate entries (allowed under the rules) and taking opposite positions (using single stock futures) in each account. The individuals total market risk is zero, but if the underlying security does move in either direction, one of the accountants could show a very positive balance and could thus win the competition.

It’s only a small step to consider creating 16 accounts, using half to make major, geared bets in one direction and the other 8 the same bet or “investment” in the opposite direction. Close out the 8 losers and proceed, as above, with the 8 winners. For a reasonably small entrance fee on each of multiple entries, investors can massively increase their chance of gaming the game, without any semblance of investment, trading or even speculative skill.

The winner becomes he most able to manipulate the rules.

Search Engine Optimisation (SEO) is an unfortunately necessary part of driving traffic to discover a website. Good content is necessary but not sufficient.

Why SEO is necessary

A large percentage of web traffic is directed by search engines. After all, this is how Google has become the giant company that it is. I tried to track down some hard statistics on this, but they varied widely and didn’t seem all that credible. Nevertheless I think it is clear that this traffic is signficant.

Search engines use algorithms and automated scripts (“spiders”) to understand the importance, quality, relevance and popularity of content on the web. A radiographer takes xrays without being able to see directly the same picture as the xray will produce. A photographer taking black-and-white photographs needs to ignore the colour in the viewfinder and imagine the light and shadows and shapes of the final photograph.

If your website has excellent content, but structures it in a way that is not readily accessible to a search engine’s spiders, then the spider will pass on by without sending humans to visit your site. Two easy examples may help:

1. Flash content – the content may look great and be ground-breaking and useful, but since most spiders don’t currently “understand” Flash content, it will be ignored.
2. Login, registration and forms – if large parts of a website are accessible only after filling information in a form or registering and logging in, the spiders won’t get in the door.

There are other considerations that are postulated to be relevant:

• Duplicate content “dilutes” the scores of any individual page
• Many links to irrelevant, poorly rated pages can suggest that your site is not providing useful info to the user. This effect is stronger since search engines try to separate “link farms” and rings and other methods to make a collection of websites appear more connected than they are in reality.
• Poor choice of keywords that searchers may often use, or targeting terms that are widely targeted by a range of other websites.

A typical SEO strategy is quite complex and takes times, effort and money

A typical SEO strategy would cover analysing the target audience of a site, understanding the site content, understanding the site structure, doing keyword analysis, checking out competitors, generating a few good quality inbound links if applicable, possibly generating some linkbait content, installing appropriate tools (e.g. Google Analytics) to monitor traffic and then repeating the cycle once the customer behaviour is better understood. Key metrics are site traffic generated, low bounce rates, long time on the site, repeat customers, higher sales (or more contacts if online sales aren’t part of your service) and higher search rankings.

All of this takes time (both from the SEO but also from the website owner). There are many fly-by-night organisations claiming to do SEO with neither the knowledge or the business ethics to get it right. It is probably because it is a poorly understood, sometimes arcane speciality, that these companies get into business with low starting capital costs.

How not to do SEO in ten easy steps

I received an unsolicited email from Zenteq recently. I’m not providing a link to their website as I have no reason to believe they can deliver anything useful.

1. Sending unsolicited email (aka SPAM). This is typically a bad idea. Best case scenario you get a few new customers. Worst case scenario you irritate a huge block of potential customers, have your mail server and/or IP blocked as a source of spam, have your ISP close down your website for abuse and so on.
2. Use bright (as in reflective safety wear) green text and truly ugly formatting. Not a professional image by far.
3. Offer to “SEO” the website by submitting to 600,000 search engines monthly. This is irrelevant and a giant waste of money.
4. Charge R350 per month. In the short-term, this is far too little. The work involved at the outset of optimising a website for search engines requires several full days of work. However, in the long-run, this may well be too much. Since there appears to be no reason for the client to stay with Zenteq, we have a familiar problem where the business model doesn’t make sense for a serious operator and thus it’s likely that it isn’t a serious operator. (on trawling their webpage I see there is a R1000 upfront fee as well. Nice not to include this in the email. It still isn’t enough for serious upfront work)
5. No description of other components of SEO strategy, or examples of prior successful work.
6. “From” email is marketing@fire-equipment.org, “Reply to” is newheights70@telkomsa.net but the content directs the reader to info@zenteq.co.za. So which is it?
7. Structure your email so that it gets stopped by the spam filter built into both Thunderbird and Apple’s Mail application.
8. Include icons on their site claiming valid XHTML code, but then fail the test when the button is clicked.
9. Analysis of google results shows no links to zenteq.co.za.
10. And my favourite – a search on google for “seo site:za” which searches for the top websites relating to “seo” within the “za” domain doesn’t have zenteq listed in the top ten pages. A first-page listing is almost an requirement if you expect any number of click-throughs.

So who guards the guards, and who optimises the optimisers?