Tuesday, April 21, 2009

Let a 1000 flowers boom - Using Income-Contigent Loans with Startups

The Treasury, WhitehallImage via Wikipedia

TechCrunch UK posted an open letter by Robin Klein of The Accelerator Group to the Chancellor of the Exchequer Alistair Darling and Lord Dryson Minster for Science and Innovation about what to do with purported stimulus funds. BVCA wants the money to go to large VC funds whereas Robin Klein wants to see the money channelled to supporting very early stage companies (amounts less than £100k).
Robin’s logic and reasoning is sound and I agree with them. But it is not a good use of the money for two reasons.

Tech (web) Focused
The idea is far too technology (read web) focused. There are lots of opportunities throughout theUK for entrepreneurs to create businesses; many, indeed most, outside the world of the web. Why shouldn’t someone starting a lawn-mowing business have access to early stage funding as a technology developer? Both create value. We in the technology sector tend to be myopic about start-ups, small businesses and entrepreneurs. Richard Branson can hardly be accused of creating a technology business and yet he is by most measures the UK’s most successful entrepreneur.

Yes, technology creates long term value and wealth, but the vast majority of wealth is created by companies outside of the technology sector using technology and not developing it. It is created by a lawn-moving business using twitter to alert their customers that their lawn is done and having a website where clients can go and book a visit using something like BookingBug to provide the functionality. The lawn-mowing business is creating value through better customer service and consequently generates wealth. Would a business angle or early seed stage fund invest in such a company? What about if it is located in the hinterlands of Wales?

Relying on Judgement
The mechanism for distributing funding relies on someone making a judgement call as to what is potentially a good opportunity. The act of making a judgement takes time and as many commentators pointed out in response to the open letter, time is very precious at the early stages of a business. Waiting more than a month for a response is a massive drag on very early stage businesses. They need responses fast.

More problematic is that a person can only make the judgement based on their experience and expertise. Many great opportunities will be bypassed as the judges’ focus on what they know. Now however is a time to fund companies that are moving into new areas and new ways. It is a time to let 1000 flowers bloom. In the end the only real judgement that matters is that of the market. It would be better to create a situation where those companies can be judged by the market rather than a limited individual. The market is crowd-sourced investment decisions.

In place of co-investing or creating lots of seed funds, I propose that the UK government create a scheme of income-contingent loans. Under the scheme an entrepreneur can take out a loan that covers his previous salary up to a maximum of £50k to £60k. The loan is paid monthly like salary and is re-paid by the individual (not the company) through the tax system (similar to student loans). Other characteristics of the scheme are:
  • The scheme would provide loans for up to 3 people per business in the first year, followed by another 2 new employees in the second year
  • The loans are tied to the individual and are re-paid by the individual based on the individuals income
  • An individual can only take out a loan under this scheme once every 5 years
An income-contingent loan scheme provides funding irrespective of industry or goods and services. It addresses the funding gap that is a barrier to entry for all entrepreneurs and has a lower administrative burden. The loan scheme can be administrated through the existing Government banks and through an online loan application system which are widely geographically diverse, scalable and most importantly can return a fast decision.

One big objection is the potential for fraud. Nothing involving money is without the potential for fraud and venture funding is not immune (witness Tiger Telematics). By putting the liability to re-pay the loan onto the individual reduces the avenues for fraud using this scheme. The other limitations are also designed to reduce the attractiveness of fraudulent behaviour.

Granted, the loan scheme is unlikely to produce the next Google but I would rather see the loan scheme generate 100,000 businesses all employing an average of 10 people. That would be far more valuable to the UK economy as a whole than 1 Google.

Ideally, you would run both an income-contingent loan scheme and co-invest in early stage investments. However, given the realities the loan scheme is more valuable. The co-investment scheme should follow. By the time the co-investment scheme is up and running many of the first lot of companies that have benefitted from the loan scheme will be ready for their first round of funding.

Reblog this post [with Zemanta]

Friday, April 10, 2009

Capturing Intention in User Behaviour - Do it Directly

Taken from Princes St.Image via Wikipedia

I had an interesting conversation at OpenCoffee London yesterday around personalisation and recommendations. At they table were several people involved in companies that are in the area of marketing analytics and personalisation and recommendations.

One interesting point raised was that as targeting becomes more and more focused, there is a loss of information that can harm personalisation/recommendation algorithms. There is a point where targeting stops being effective in delivering relevant personalisation and recommendations. I'll return to this later.

Of more note was how capturing attention is difficult, indeed impossible with behavioural tracking. In "Wither Social Networks, Arise Communities" and "Resources vs Answers - Asking a Question of Search" posts I've look at the importance of intent to properly target a user. What came in the conversation yesterday was be direct. Essentially, ask the user what their intent is and stop trying to guess. Allow the user to tell you (the business/software etc.) exactly what their intent is.

The example used was booking flights to Edinburgh. Currently travel sites get you to input the departure, destination and dates. Then give you lots of results which you have to wade through and links to ads for various travel related items in Edinburgh. However, there is no way for the site to tell what links it should show or how to order the results.

If however, the user starts by telling the service "I want flights to Edinburgh for a weekend of fun" suddenly the service has a slew of information that it can use to target the right ads and services and also organise the results. From that simple statement, the system "knows" the following information:
  • The flights need to be Friday afternoon/evening and Sunday afternoon/evening
  • The purpose is fun so flights need to be cheap/reasonable
  • The person will be looking for tourist, entertainment stuff to do
From this new knowledge the system can prioritise flight results and target service suggestions and advertising better to the person's intent. This wouldn't be hard to do. It is a step away from normal form-based UI into NLP but it opens a whole new vista of opportunities.
Reblog this post [with Zemanta]

Wednesday, April 08, 2009

Open Access Will Make the NBN Work, not Price

Ok this is getting ridiculous. There is lots of commentary going on in the Australian press and from the politicians about the NBN. Most of it inane and demonstrating the commenter's complete lack of understanding about the technology and the issue.

The NBN will provide economic, social and environmental benefits. This fascination and indeed obsession with cost of access is self-defeating. Here is a quick overview of the benefits

  • Economic: NBN improves access to SaaS solutions and increases effective collaboration both improving producing and reducing costs. NBN opens up new possibilities for new companies and industries that hadn't existing before. Could anyone have predicted YouTube or Twitter when we were using 56k modems? NBN brings very high speed access to businesses of all sizes rather than just the largest.
  • Social: NBN, as others pointed out, provides the ability to role out education and health applications right to where people are. Some are already talking about providing free basic access to the poor.
  • Environmental: NBN will create a boom in video calling and telepresence cutting down on travel and reducing pollution. The NBN goes a long way to reducing energy intensity of the economy.
The NBN is the convergence of communication and data. It will be used for applications ranging from video calling to web access to television and radio to smart metering to health monitoring. To see it only in the light of access misses the larger value of the NBN. Nor is the speed a big issue either. Speed through the fibre is determined by the equipment at either end and not by the cable itself. Should users and the access providers decide they can put in the equipment that supports 1Gbps without much problem. Far more than copper, fibre is future proof.

Have a look at these videos that Microsoft produced as a look into the future of computers. The NBN is the key to enabling this world.

The key to the NBN living up to its promise is how the Government structures it and it is quite possible they will get it wrong. That shouldn't stop us from trying though. Life is a risk and we cannot be afraid of making a mistake.

For the NBN to fulfil its promise the following items have to be done:
  • The conduit/real estate and fibre should be help in separate companies
  • The company sold at the end of the 5 years should only be the fibre layer and assets. The conduit/real estate company needs to remain majority Government owned.
  • The legislation has to guarantee Open Access to both the conduits (so other companies can label cabling) and the fibre.
  • Fibre should be laid direct to the house. The NBN shouldn't use shared access.
  • Access must be symmetric the same upload speed as download speed.
By embedding Open Access to the conduits and the fibre you create an environment where organisations can compete on the fibre but also at the physical layer. The largest barrier to entry remains digging up the ground and laying conduit followed by the fibre.

The Opposition needs to hold the Government to implementing the about principals. They have to work with the Government to split the construction of the NBN into two parts:
  • Building the Conduits. This is the stimulus part of the NBN and can begin quickly and should start in the large regional towns before moving to the city
  • Laying the fibre and installing the Equipment. This is the key planning stage and in order to avoid prolong delays, can be done will the conduits are being built.
The conduit building can be done by any construction company that can dig a trench, lay piping and install access points. The contracts should be standard and let out to as many of the construction companies as possible. There is no need at all for one company or small group of companies to win the contract to build all the conduits. In terms of speed by letting out contracts to various companies around Australia the construction can go on in parallel in many different locations.

The core issue here is not the price of access. It is Open Access. The Opposition must be holding the Government to providing Open Access not only to the fibre but also to the conduits. Open Access is the key to the NBN working.

Reblog this post [with Zemanta]

Tuesday, April 07, 2009

Busting Recessions - Income Contingency Loans for Redundancy & Training

I’ve touched on income-contingency loans in previous posts, time to explore them in more detail.


The point of redundancy is to allow businesses to shed staff as necessary without throwing the people onto the garbage heap. The flexibility provided to hire and fire staff is important in keeping an economy resilient and flexible. Society also benefits from redundancy payments as it reduces the fear of not being able to pay bills. As most redundancies occur during recessions it helps to cushion the demand destruction of job losses.
There are two problems with redundancy systems as they currently stand:

  1. From the point of view of a person, redundancy payments are not enough to survive on, and
  2. From the point of view of the company, redundancy payments are a big hit to cashflow.
The Nordic countries use generous welfare system to reconcile the two competing interests but have high taxation as a result.

The same reconciliation of the competing interests can be achieved by using income-contingent loans. When making someone redundant the company takes out a loan from the Government to pay for the redundancy of the person. The loan is then paid back through the taxation system from the profits/income of the business taking out the loan.

Redundancy payments can then become 75-90% of final salary for a period that can last 6 to 1 year without driving businesses into the ground.

Using income contingent loans allows:
  • Businesses to lessen the impact on cashflow from redundancy payments
  • Individuals get larger payments that reflect their final salary for longer periods
  • Redundancy can be provided to all employees instead of just those that have been at the business for a long time
  • Bills, rent and mortgage continue to be paid reducing negative flow on effects from redundancy to the wider economy
  • Redundancy is guaranteed even if the company eventually goes bust

Training & Education

One of the aspects of recession is that there are too many people chasing too few jobs. Income-contingent redundancy helps lengthen the time an individual has to find another job but it isn’t the only way of smoothing things out.
The second method is to provide income-contingent loans to businesses for training employees. The business would take out an income-contingent loan that would pay for the salary and training. A modification is that the loan is jointly held by the individual and the business.

These loans would achieve:
  • Improved cashflow for the business as the employee is effectively removed from the payroll
  • The person is improving the skills and training so that they are more effective when business picks up
  • The business can hold onto good employees that would otherwise have to be made redundant
  • Less people are chasing the limited number of jobs
  • Bills, rent and mortgage continue to be paid reducing negative flow on effects of redundancy/job losses to the wider economy
  • Society benefits from better trained and skilled workforce
Income contingent loans better address the needs of business and employees without trying to jam them all into the same one-size-fits-all. They recognise the benefits of redundancy payments and training doesn’t accrue just to the business or individual or society but to all three in varying amounts.

On the face of it the training loans are better overall but training will not be suitable for all situations or occurrences. Both systems should be implemented to allow flexibility and support the uniqueness of each redundancy situation.

Reblog this post [with Zemanta]

Aussie Gov re-entering the telco game

Optical fiber provides cheaper bandwidth for l...Image via Wikipedia

The Rudd Labour Government has announced that rather than tendering building fibre optic broadband network to a private consortium or telco, they are going to do it themselves. At a cost of AU$43bn over 8 years.

The logic sounds nice nice on-going stimulus during a time of economic crisis that is productivity enhancing. Broadly I agree with the sentiment. Broadband is a viable stimulus that will have positive benefits in productivity. The larger the pipe the greater the economic opportunity and productivity.

I have two problems with the announcement though.
  1. It does seem like an idealogical policy to get the government back into the telco game - not something that is warranted or needed
  2. More importantly the value of the Government involvement is limited
Where the Government should be involved is creating a national conduit system which is open and easy for any company or organisation to run their cables. The largest cost in building a broadband network is digging up the ground and putting in the conduits to carry the cables. By doing this the single biggest barrier to entry into the provision of broadband is removed which will do more to increase competition and improve services than all the regulation and legal proceedings.

Laying their own cable is an unnecessary expense, but given that this is likely to happen then I hope the Government learns the lesson of KPN in The Netherlands that open access to the network benefits not only the consumer, but the owner of the network. Open Access has to be a requirement for the Government network and must be a requirement if the network is ever sold to the private sector.

Reblog this post [with Zemanta]

Monday, April 06, 2009

Saul Griffith, Energy Intensity and Trading Negawatts

[ebook] A History of the Science and Politics ...Image by Changhua Coast Conservation Action via Flickr

Saul Griffith of Squid Labs fame recently gave a presentation at Green:Net looking at the engineering challenge that is climate change/energy renewal. It is an informative presentation and one that builds on previous work he has done to quantify the problem. However, since seeing the math on a BBC documentary and then again in this presentation something has been bugging me and I couldn’t put my finger on it.

Saul certainly produced a fine statement of the engineering problem. Saul isn’t the first to layout the problem in that way David McKay of Cambridge University is doing a more comprehensive expose of it. Either way it is a sobering presentation.

Considered in the context of the system, though, I think some to the base assumptions may be wrong. I expect that the carbon reduction is not a linear progression but will grow slowly until a tipping point is reached at which point the carbon in the economy will drop seemingly overnight.

More importantly, the focus on carbon is leading the world astray. The primary problem isn’t carbon per se rather the energy intensity of the global economy is growing. While global energy intensity per unit GDP grows the sources of energy is relatively academic. The core problem remains the growing energy intensity per unit GDP. This is what has been bugging me about Saul's presentation - it was from the context of carbon and not about energy intensity.

To demonstrate what I mean consider Peak Oil. The Peak Oil problem is only a problem above a certain energy intensity of GDP. Below that level two effects are seen, 1) the life of oil reserves increases and (more importantly) 2) substitution for other energy sources becomes easier. By reducing the amount of oil used by reducing energy consumption less carbon is produced. The advantage is that reducing is far easier than substitution.

The answer to climate change is to set in place incentives and processes that reduce the energy intensity faster than the economy grows. Energy consumption can only be reduced when saving energy is worth more per kW than consuming energy. Some ways of doing this are:

  • Net Present Value in Energy for energy generation schemes: Like Net Present Value for evaluating monetary investment this looks at how much energy is needed over the life of the generator. All generators use energy in order to produce energy. Using NPVE will allow the comparison to see how different generation schemes reduce overall energy intensity.
  • Energy Audits: The audits would show businesses and consumers where the energy is going and suggest real changes that would make material differences to the energy they use and consume.
  • Stamp Duties/Sales blocks for low efficiency housing: Energy efficiency is in for homes with steadily rising standards but new homes only make up a small percentage of the housing stock. The biggest impact in home energy consumption will be made in improving existing housing stock. The only way to incentivise owners to improve energy consumption across the board is to either not allow the sale/rental of homes that don’t meet a certain energy standard or add an energy duty onto of the stamp duty that is determined by the energy consumption of the house.
  • Life-cycle energy labels: Transparency works wonders and the more relevant information consumers have about the products and services they consume the better purchasing decisions they can make. Providing a single number that shows how much energy is tied up in a good or service will help consumers pick products and services that consume less energy over their life and so reduce energy intensity of the economy.
In the final analysis reducing energy consumption will only be self-sustaining when the negawatt is tradable. Ultimately, personnel energy budgets or some sort of cap and trade system will work best. Trading negawatts is far more tangible than trading carbon. Carbon is intangible and there is no daily interaction with it. People understand energy; they use it every day and can see its effect. That single fact will see the far more rapid adoption of energy/negawatt trading than carbon trading.
Reblog this post [with Zemanta]