Having been consulted by African and Latin American authorities, I came to the conclusion that there are a few common elements to set up an incubator/accelerator for new businesses. Here are the most frequent elements that I wrote down…
i) Public initiatives work only if they are here to coordinate, support and create a critical mass of attention. They should not set priorities, directions or imply any other kind of active involvement, as this is counter-productive. Countless examples of public initiatives (including France, Malaysia, but also the EU) tried to replicate the Silicon Valley by setting strong directions on certain industries. This did not succeed.
ii) Public initiatives work mainly when they provide a clear framework for selecting start-ups, providing them with a clear time-frame of support, and with specific means answering real needs. Assessing these needs is the key success factor to succeed in the project. These needs vary depending on the industries covered, the maturity of companies, the location (country/city) and the competences gathered by the company.
iii) Certain needs can be served by the private sector. The public initiative should not compete with and crowd out the private sector. A key success factor in the Silicon Valley, in London, in Berlin and in Paris has been that more than start-ups, it is an ecosystem of private providers which has been built, with a critical mass of actors helping and nurturing start-ups. The public initiative can filter our the good providers from bad ones, and apply a quality level, as well as negotiate prices for the start-ups.
iv) Private initiatives can provide cheap location, high level equipment, certain mutualised services and backing. Backing can be financial, but should not be in money – preferably as a « checkbook» of services (time of lawyers, accountants, IP specialists, etc.). Money/grants/loans should be a small portion only and used as a « signal» only (otherwise, entrepreneurs come, take the money, and then leave to the US… like in Chile).
v) The partners of the public initiative should at least be: business angels networks, non-profit such as Endeavor who mentor entrepreneurs, VC firms, large firms, banks, but also headhunters, law firms, accountants, outsourced marketing specialists, the stock exchange (if possible to facilitate potential IPOs in a pilot program).
vi) The public initiative should at least fill two gaps: the relationship between labs (private, University, public) and entrepreneurs; and large firms and entrepreneurs. First because it is necessary to identify early innovations (possibly by developing a partnership with business schools to have Masters students work on that on the template developed by Aalto or HEIG-VD), and then identify the key HR needs (notably experienced executives). Second, because a key problem of start-ups is to find their first customers. Large firms are difficult to get access to, and are good beta-testers and first references.
vii) The public initiative should focus as much as possible on internationalization of start-ups. Hence, it has to take the risk of seeing some start-ups leave and reincorporate in the US or elsewhere. However, a key success factor to develop viable companies is to attract stable foreign capital. Incorporating start-ups in Cayman, Delaware or other « neutral» places is part of it (managing the real exchange rate being another problem, until now not solved). This means that OPIC, the Interamerican Bank of Development, and other foreign groups should be contacted, and some « tours» should be organized for selected start-ups to visit key countries and meet local representatives (financial sources, clients, suppliers and help to open branches).
viii) Specific needs might entail offering classes to help entrepreneurs speak fluent English, communicate according to IFRS/International GAAP/US GAAP, outsource some functions abroad as they are cheaper.
Another paper published on the SSRN, dedicated to fees, performance and alignment of interests in private equity (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2296971).
Here is the abstract:
While PE data is subject to time-lag and still US-centric, twenty years of data provide grounds for a thorough analysis of private equity funds (PEF) performances. To do so, we model gross returns for US and EMEA PEFs. We benchmark returns with total market indexes (TMI) using the PME method with the DPI as a regulator of distributions. We find that average US funds perform in line with the calculated benchmarks on a net and gross basis: funds have marginally better IRRs; indexes better multiples. Carried interest has no material impact on the relative performance of funds: when a vintage underperforms the index on a net basis, this holds true on a gross basis. Top quartile funds show an outperformance on a net and gross basis, and timing of cash-flows explains part of this performance. Computing “Bain Capital scenarios” show that the 1%-30% formula is marginally more attractive than a 1.5%-20%, increasing returns by 2.3% over 16 years (0.15% per year). We conclude that more than levels of management fees and carried interest, the level of preferred return rate might reduce the alignment of interests between fund managers and investors. Calculating a spread with PME -DPI index and sharing the resulting alpha might prove more efficient.
I published on the SSRN network a paper on « The predictive power of the J-Curve» (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2296967).
Here is the abstract:
Dealing with a recurring low level of data quality in private equity, we propose a novel approach of understanding the behavior of private equity funds (PEFs): we use PEFs’ illiquidity as a factor of analysis. To do so, we measure the distance between PEF cash-flows (“J-Curves”) and return categories (“ideal-types”) that we have identified through our novel reasoning. As a result, our model excludes the attribution of a given fund from certain return categories in early years. It then attributes a fund to a specific category with a high level of confidence. By doing so, this model could help reducing solvency costs of investing in PEFs, as well as support the analysis of existing PEFs either by their current investor or for new investors on the secondary market.
(this paper was originally published on PEHub – http://www.pehub.com/154773/pe%E2%80%99s-achilles-ego/)
(this paper was originally published on PEHub – http://www.pehub.com/144359/beyond-the-model-t-of-pe/)
Anarchist for Pessoa, conservative for Ibsen, greedy for Dickens, or sometimes philanthropist, we have discovered with the LIBOR fixing scheme that the figure of the banker can be simply a fraudster. However, the banker is also the custodian of the monopoly of creating money, and for this he holds a great power. It is time that he shares his portion of the collective burden, and assumes an explicit social function.
The banker is the direct supplier of money, a power delegated to him by public authorities via central banks. This power, and the public insurances which go with it, is inseparable of certain responsibilities. Still, as much has his role has grown in importance, has diversified and increased in complexity, his responsibilities have gradually been diluted. They have become more blurry in their attribution, in their assumption and in their control as rogue trading and now LIBOR fixing have illustrated.
Technological tools have accelerated the dilution of responsibilities inside the banks. These tools have also increased the fragmentation of processes and functions inside banks. This phenomenon, has increased the difficulty to have a precise, exact and clear overview of the activity of a given bank. The banking activity has largely become bureaucratic, technocratic and impersonal.
At the same time, technology has helped banks to rake in productivity gains and to develop financial innovations. At the scale of individuals, the adoption of technological tools has resulted in an important transfer of the work from the bank towards the client. The most obvious gains of the computing and information networks revolution have been captured by the banks. They have as a counterpart to support the responsibilities which go with this gains. In that respect, there should be no free (technological) lunches for banks.
One of the responsibilities of the banker is to educate the client, as the burden of the banking relationship is on the clients’ shoulders. Education, and not marketing for the bank, sometimes under cover of “libertarian paternalism” and “nudging” which was developed by Richard Thaler, Professor in Chicago. This education is not only necessary, but it goes along the right for the client to have a clear and objective information, of which he can make sense and understand.
Education of clients is a central piece of the banking puzzle which is missing today, either in developing or developed societies. Experiences lead in Australia have nevertheless shown that a program to educate fragile retail clients to manage budgets, credit and savings have notably improved their financial situation, but also their quality of life and income level. This example should not be limited to retail clients, but should also cover all the aspects of the banking relationship – meaning that all clients should get access to education, especially when it comes to complex products and financial innovations.
It is necessary that bankers support this education, not to sell more to their clients but to sell better to them – for the public good. This would have the advantage to improve the image of bankers and to support society at large – society which has given them their power and their function. The risk associated with this education endeavor is that there could be a temptation to misuse it to support sales efforts.
To avoid this potential drift, inspiration could be found in some of the duties imposed to other intellectual professions such as attorneys-at-law. Attorneys have to provide a certain number of hours pro bono in certain countries, or to help difficult cases for very symbolic compensations. It is part of their status, and professional organizations make sure that they support their share of the social burden. Bankers would gain to get inspiration in this model, by instituting for example the equivalent of “financial peace corps”, to educate for free and for the public good.
This pro bono time could benefit to other sectors, such as micro-finance in developed countries. One of the major obstacles of the development of micro-credit in the Western world is the cost of human resources to work on these projects: its is far higher than the absolute return of the small loans granted. Pro bono time could also help regulators and third parties to get up to speed with the latest developments in financial innovations.
The biggest hurdle is to set a balance for this mechanism. However, if some companies such as Google authorize (officially) their employees to spend 20% of their time on individual initiatives (though in connection with their work), it is logical that bankers can spend an equivalent time to the benefit of the public good. This could paradoxically improve their knowledge of clients, and finally help to redirect financial innovation to the real economy.
An obligation to educate transformed in an opportunity to learn. This could be an illustration of the banker anarchist of Fernando Pessoa, or the fourierist banker of Charles Fourrier. The latter undertook to elevate socially blue-collars by providing free evening classes opened to all about sciences, such as astronomy. Educating the masses to finance might be one of the most socially disruptive endeavor which could be undertaken by bankers – whether anarchists or not.
(this paper was published initially on PEHub: http://www.pehub.com/176010/is-enemy-fact-us/)
Among the multiple reports, surveys, white papers and other contributions that regularly hit the shores of the private equity world, the Kaufman Foundation’s « We have met the enemy…» (http://www.kauffman.org/uploadedfiles/vc-enemy-is-us-report.pdf) was welcomed with an astounding silence. As my first book publisher told me, it is either a sign that it is really bad (which might have been the case of my first book, but doubtful when it comes to the Kauffman Foundation as an author); or that it is touching such a delicate subject that it becomes untouchable. I tend to think that the latter explanation is the most likely.
After all, it not only criticizes GPs of venture capital funds, but also the LPs (category to which the Kauffman Foundation belongs to). By aiming at the GP and the LP community, the Kauffman Foundation makes few friends – even though its report is most welcomed and salutary. Having practiced frankness and honesty many times myself, there is one little lesson that I learned though: what is written has to be irreproachable. If not, then the criticized take any detail or inaccuracy to cast the shadow of doubt on whatever else is written. I must say that even if I approve the main ideas of the report, it may not be accurate – and this might hurt the cause it supports and the message it conveys.
1. The « golden age» of VC will not « come back»
One of the critics that might be addressed to the report of Diane Mulcahy, Bills Weeks and Harold Bradley is that it assumes that everything remained (and should remain going forward) equal in the venture capital world. This is a problem.
On one side, the report states that the economics of fund management should reflect a better alignment of interest, by switching from the « 2%-20% model» to a pre-defined budget. On the other side, the authors expect the VC industry to generate 2x the capital invested and a « 3 percent to 5 percent annual returns above the public markets» . At the same time, the authors hope that the sector will come back to a pre-1995 situation in terms of amounts collected and invested.
Let’s make here a prophecy: the « golden age» of VC (if it ever existed) will never come back – and the Kauffman Foundation, which supports the development of entrepreneurship, should actually celebrate it. Entrepreneurs today find more readily capital (supposedly, given the amounts raised and deployed). Entrepreneur has become a job (» serial entrepreneurs» prove it), not a marginal activity for hotshots. Successful entrepreneurs are even a source of investments (» super-angels» ).
There is no turning back, because the VC industry has professionalized and it has become part of the asset allocation of institutions (an approach that the report criticizes heavily). What does it mean concretely? That the 2x capital and 300 to 500 basis points above market returns are no longer a reasonable target.
As much as VC is known, understood, explained, analysed, documented, structured… it becomes a less risky activity. From an « exotic» activity (such as collecting fine art, classic cars or wine), it is now part of the private equity asset class (itself part of alternative investments). Less risk means lower returns, due to the law of the decline of marginal returns (which applies here as much as anywhere else).
Does it mean that the 2%-20% is untouchable? No, and the authors rightfully complain that this inheritance of the early days of modern venture capital is no longer justified. However, even if the management (and other) fees are going down, they probably will never make it up to the loss of returns.
2. Right sizing: beware of cutting too close to the bone
Let’s assume for the moment that the asset allocation of institutions remains the same, and that the amount of capital available is going to drive/maintain low marginal returns in US VC. Does it mean that it is not worth investing in it? One of the conclusion of the authors is that an arbitrage with the Russell 2000 index has lead the Kauffman Foundation to reduce its exposure to some fund managers and go for ETFs replicating this index.
That might first reduce the portfolio diversification of the overall Foundation, which is an important aspect to consider. Beyond this, it is particularly uncomfortable to hear a comparison between listed and non-listed markets. The Russell 2000 is a passive index based on an exception in the financial world: the stock exchange. The authors compare the activity of trading listed shares on a secondary market with the capital increase in early stage companies. Methodologically, that sounds shaky because it assumes that managers in the VC world just pick stocks and shares as an index builder would do and an ETF provider would operate. That does not seem really the case (everyone who tried to negotiate a shareholder’s agreement with the management of a portfolio company would agree – and this is a small portion of the job).
Does it mean that GPs of large VC funds (» 1 billion+ USD» ) deserve their fees? Not necessarily. However, when the authors emphasize the necessity to focus on funds with less than 400 million USD under management, the logic is that these management fees will not necessarily go down dramatically. One of the reasons is the fact that doing the job actually requires increasingly more money (and not only to pay hefty salaries and bonuses as the authors seem to assume): compliance with regulations is more demanding, as well as faster and more detailed reporting. This requires manpower and resources.
By going to low on the management fee topic, the authors might risk to push the GPs either to seek their compensation elsewhere (out of the LP radar, for example, by charging the portfolio companies) or simply to cut corners. Pushing the fund size down might have another consequence: having funds with a sub-optimal size (with no critical mass to amortize the fixed costs). Here, an example comes to mind: European VC, which has even more dismal returns than US VC. Part of the reasons are notably that the average fund size is rather low, which means that fees and expenses are actually higher than for their American counterparts.
3. The J-Curve phenomenon exist: I have seen it
There is a strong attack against the J-Curve theory in the report. Inexplicably, the authors assume that the J-Curve is the reflection of the projections of successive IRRs over the life time of the fund. This is totally wrong: it is the projection of cumulated cash-flows of the funds. Hence, not only the J-Curve exists, but this is a permanent phenomenon that can be seen by looking at any private equity fund.
The authors build on this misunderstanding of the basic phenomenon to criticize the « seductive narratives» build by GPs willing to raise the next fund. Said differently, IRRs are inflated by GPs to fund raise and are not reflecting the true value of the funds.
Nobody defends IRRs here, but the authors are probably aware that the valuation guidelines are clear: either a portfolio company is kept at cost by a GP, or it is depreciated if the portfolio company is under budget, or it is appreciated if there was an upward round (ie, a third party made the new valuation of the company). Except in outright fraud, GPs usually apply these guidelines (and the valuations are audited every year by external auditors).
What to conclude of the « n-shape» curve that successive IRRs of VC funds show?
i) That the authors should use it to challenge the TVPIs of their funds and not to take an investment decision (which is what they conclude at the end of the report).
ii) That their GPs invest in companies that could raise further (upward) rounds. Is it bad? Not necessarily, if the alternative is a write-off of the said portfolio companies. Yes, if the upward rounds are the proof of a bubble, which deflates with the trade sale or a mediocre IPO.
This is actually a consequence of the « too much capital chasing too few cherries» that the authors deplore. That does not mean that the J-Curve does not exist, and that the actual performance of fund, whether the authors like it or not, cannot seriously be analysed before year 5 to 7 of a given fund’s life.
4. Praising evergreen vehicle: beware of what you wish for, it might come true
Against the standard 10 years lifespan of a fund, the authors praise the evergreen structure. It « reduces the impact of cumulative fees and eliminates the time pressure to produce positive returns in time of the next fundraise» (p. 31).
First, it is not clear how an evergreen vehicle might reduce the impact of cumulative fees. Actually, evergreen structures are more of a « black box» than the typical LP-GP relationship. Not only investment and management money are blended, but there is a lot to criticize about the lack of transparency of listed evergreen private equity vehicles (hence their structural 20 to 30% discount to NAV).
Let’s compare the two:
i) limited parnterships: management fees limited partnerships are usually calculated on the capital committed (sometimes invested) during the investment period, and then are winding up during the divestment period (and are calculated either on the NAV of the fund or according to alternative formula). The authors fail to recognize this basic rule in their table page 34 when they assume that the management fees are fixed and equal over the 10 years life of the fund.
ii) evergreen vehicle: because there is no investment/divestment period, the fees are calculated as a percentage of the assets under management and usually not obvious to identify. They are in the P&L of the structure and not negotiated with the investors (at least in the listed vehicles). This means that investors in evergreen structures pay more on the overall life of the structure, but also that they actually do not know how much they pay at any given time – and that the management of the evergreen vehicle can increase its compensation at will.
The « time pressure» associated with limited partnerships is actually… a good thing. The worst which can happen to an investor is when a GP falls in love with its portfolio companies (or worse, uses them as personal cash cows to charge « advisory fees» for example). The ten year’s time is necessary to force the GP to manage actively its portfolio, to create value and to give the opportunity of LPs to « exit» (if not « voice» their discontent). LPs in evergreen structure have no « voice» (being divided), but they cannot even exit (or in worse conditions than with a standard limited partnership stake).
As a side note, the authors mention an exit window every four years (page 38) which is an even shorter time frame than the usual ten years of a limited partnership. It is difficult to criticize the « short-term approach» of GPs in a standard limited partnership and in the same breath explain that an evergreen vehicle with four-years exit windows should be the solution.
Last but not least: limited partnerships are tax efficient – and distribute profits. Evergreen vehicles do not distribute (they reinvest) and if they do, they distribute dividends (which are not tax efficient).
5. Terms and conditions: the grass is always greener elsewhere… until you get there
The authors criticize the terms and conditions of limited partnerships and the practices associated with it:
i) deal-by-deal carried is (rightfully) excoriated. The fact is that it is more and more an exception. First it is a US-only feature. Then even in the US, only the top VC fund managers can still impose it. So complaining about it is like complaining that this GP is making too much money. That’s why we do not have clawbacks in Europe: we do not have deal-by-deal carry.
ii) the lack of preferred return (or « hurdle» ) rate. Once again, this is an exception. Almost all European funds have a hurdle rate and only the top US VC firms can skip this feature…
iii) serial fund raising « every twenty-four to thirty-six months» . First, it is a good thing (exit possibility for LPs and keep GPs on their toes), and then in Europe and increasingly the US, there are provisions to prevent fund raising before the GP has invested 70% or more of the current fund.
The limited partnership is not perfect, but at the Luxembourgers would tell you, it is difficult to do better: the SICAR took them ten years to be engineered and is still not very popular compared to a typical LP. Luxembourg even created the Special Investment Fund to get closer to the LP model… Of course, everyone is listening if there is a better solution – but that will certainly not be an evergreen fund.
This report is full of great ideas and suggestions. It is hence more annoying to find the elements listed above in it. Does it mean that the conclusions of the report should be overlooked? Not at all, but the authors should have gone the extra mile to give their report (and its conclusions) the magnitude that it deserved. One simple example: it is great to support the PME method to benchmark VC funds. However, the way it is applied in the Exhibit of the reports falls under the same critics which were addressed to Kocis & alii, as well as Schoar and Kaplan about the difficulty to deal with selling the index. Had the authors made a bit more research, they might have come across the PME+ (developed by Rouvinez) method which solves this major hurdle. The same applies to the comments above…
A common belief about humanity is that it can learn from its past mistakes. The Newtown shooting, beyond its emotional consequences, is a collective mistake. To find prevent a Newtown or Columbine tragic event to happen again, the public attention is focused on personal weapons control. Seen from Switzerland, where basically every single male national has a mandatory (military) gun at home, and no school shooting to report for, it is a puzzle: American deadly shootings trigger again and again the same debate, without virtually any progress.
The reason might be that finding a solution requires to approach this problem from a different angle – or even to looking at it indirectly. This approach has lead to discovery the Higgs Boson (the “God particle”), as physicians looked at the result of the collision of particles to find traces of the Higgs. Applying this intellectual approach to the shootings question might produce interesting results.
Putting mass shootings (a social regression) on a collision course with what embodies the American notion of progress (the Silicon Valley) sounds a bit odd. However, the result would probably be more surprising: one of them being very cheap guns produced at home thanks to 3D-printing. No record. No authorization. Not even a valid possibility to estimate of the number of weapons circulating on American soil – hence a regression. Just like drugs (such as cocaine or heroine) control programs failed to take into account the US DIY drugs (such as metamfetamin), weapon control might be defeated before even seriously starting by anyone with a computer (500 USD), a 3D printer (2000 USD) and templates downloaded for free on the internet.
School shootings and Silicon Valley have other intersections, namely externalities. Shootings have happened first and foremost because young individuals have regularly been rejected by the grinding American high school system – and the American society at large.The psychical situation of teenage shooters, though a recurring debate, is not the primary matter in that instance. Indeed, nowhere else than in the US individuals would bear the full responsibility of such events without society questioning its own underlying structures.
However, the US do not question their extreme competitiveness, their rejection of “failed” individuals and their inability to cope with “the margins”. The current education and social system have lead companies to complain that young Americans are increasingly difficult to manage, to train at working in teams and to keep in a given company as loyal employees. They also have trouble to deal with less successful individuals.
There is hence something wrong with American values currently taught to youngsters. Said differently, the American society for all its progress, has reached a certain limit in terms of social development. These shootings are a proof of it. Weapon control hence appears as just a side – and somehow minor – question. Of course, questioning some of the bases of the American society is much more difficult than debating arms control. This probably explains why so far no-one dared.
One could object though: why is Silicon Valley successful if not because of these self-centered, individualistic and success-oriented entrepreneurs? Because the most successful model of entrepreneurship is built on teams of entrepreneurs (Jobs & Wozniak, Page & Brin, Gates & Allen, Hewlett & Packard), who capitalize on extended given (social, cultural, ethnic, religious) or built (friends, venture capitalists, fellow entrepreneurs, business contacts, University alumni) networks aiming at cooperation (with venture capitalists, key recruits, pilot clients, adventurous suppliers).
Moreover, successful entrepreneurs in IT are often marginals, described as “nerds” or “geeks”. They might suffer for example from a mild or severe form of Asperger syndrome, with the social consequences that go with it. Does it make them less suitable to succeed? To live in and be part of a society? No.
The American society seems to have yet to learn that differences, including (social) disabilities, are a source of richness. Unless this is undertaken, the shootings will remain recurring tragic events. It also means that what is the source of a major innovation center would irremediably fade away.
My interview on « Intégrale Placement» following the publication of the 4th edition of « Introduction au private equity» (RB Edition): http://http2.infocast.yacast.net/player/player.php?idsequence=214012&sign=387db176937d13a48f4c1e0e553b2855