Monday, March 28, 2011

Appropriate Compensation #1

For a few weeks in February and early March there was just a hint of a Buyers’ strike in the TRE auction dynamic. Many more auctions stayed open for hours or even days. Fewer buy-out bids were made. More Sellers adjusted pricing upward to close deals.

That hasn’t completely ended. We’re still seeing some egregious over-reaching by some Sellers and some push-back against that by Buyers.

One relatively new Seller this past week wound up agreeing to terms roughly double its initially-posted buy-out level. That says more about the unrealistic ask-price than the final auction terms, though.

The average returns to Buyers during March have fallen back from their brief February rise and it’s possible that we’ll close the month with new lows in average market-wide returns.

It’s not fair to say that TRE auctions are “priced for perfection”. They’re not.

But it IS fair, I think, to ask whether they are priced appropriately in light of the risk assumed by the Buyers.

So this is the first in a series of posts examining the question of what might be an appropriate level of compensation for risk in TRE auctions.

I think it’s useful at the start of this conversation to acknowledge that it can be perilous to try to analyze ANYTHING at the level of an entire market or to use market-wide averages as benchmarks. Compensation for risk is only one component of the total return required by investors, whether they are Buyers on TRE or involved in any other investment activity.

Cost of money, cost of operations, applicable taxation levels and an appropriate net profit also have to be considered.

If we were to look at the traditional factoring community we’d find that these individual variables tend towards a common level reflecting the similarities among the participants in a well-established industry. That would be true in any mature segment of the investment community. The economics of participants in a mature market tend to be shaped by the market over time and to converge around parameters that reflect the investment characteristics of that market.

That is not currently the case when we look at the TRE Buyer community. It is nowhere near as homogeneous a group as is the traditional factoring community or the community of firms that specialize in any other mature investment market.

In fact, the TRE Buyer community is an extremely diverse group. My guess is that there are substantial differences within the group in very basic characteristics like cost of funds and cost of operations.

And the motivations of the TRE Buyer community, as reflected in the determination of what might be an “appropriate” level of profit, are just as diverse.

Some might be simply looking for a place to “park” short-term funds at an expected rate better than the near-zero current money market returns. Their TRE activity is a side-line, at best; maybe even a short term experiment.

There are other Buyers, though, that are looking at TRE as potentially a primary business; one that has to both cover reasonable costs and generate reasonable profits.

So the total returns considered adequate by TRE Buyers will probably fall in a wider range than would the returns of either a fairly homogeneous industry or of a well-established asset class. TRE really represents neither of those at this point.

On the other hand, TRE DOES represent a closed system when the question of assessment of investment risk is concerned. All who choose to participate in the TRE market assume the risk that the market presents. They will react to that risk in different ways: i.e. via diversification rules, Seller-quality rules, auction-characteristic rules, etc. But they are nevertheless participating in a market that has some common risk parameters.

Our ability to define those parameters is limited by experience. TRE presents a new approach to its market and its history is short.

But some experiences external to TRE can be used with value to analyze risks that are specific to TRE. Because of the relative lack of experience with and information about TRE-specific risks I think we actually HAVE TO look outside of the limited experience of TRE itself if we are to have a meaningful discussion.

In some cases we’ll be able to offer quantitative data drawn from other sources and markets. In many we’ll only be able to suggest relative measures: e.g. that a certain TRE-specific risk is likely more than or less than that of an alternative.

But even if we can’t offer actual quantitative measures, it is still useful to consider a particular source of risk and to ask in some disciplined way whether the risk facing a TRE Buyer is likely to be greater than or less than that faced by participants in other markets that present similar challenges.

I don’t know yet how many posts this subject will occupy but I suspect it will be several; maybe half a dozen. So there’s plenty of opportunity for readers to help shape the conversation.

I welcome any suggestions of issues that should be examined in this conversation. And I’d be delighted to receive any information, especially good data, that anyone might be willing to share.

One of the things we’ll need to address, for instance, in establishing a relative benchmark, is the loss experience of the traditional factoring and receivables-finance markets. Some data on that is public, of course, but any good information that might be shared on that or any other relevant issue would be much appreciated.

We have to acknowledge at the beginning of this exercise that its primary value might be in the exercise itself as opposed to the conclusions. But that’s OK, too.

We’ll take value where we can find it.



    Sir - there is discussion regarding what appears to be some pretty large frauds that have been perpetrated by Sellers on the Exchange, I would be interested in hearing your thoughts about this.

  2. There have been two widely reported cases of alleged Seller fraud against TRE and its Buyers. Both of those cases are still in litigation, as far as I know, and we don't know what the ultimate outcome will be.

    But your point is well taken.

    Examining the extent of likely losses from purchases of TRE-auctioned receivables is exactly the point of this series of posts.

    Fraud is one of the most common sources of loss in the factoring business. If you spend enough time in the business you're highly likely to experience fraud loss.

    The issues are: how to minimize those losses, how to anticipate their magnitude and how to analyze the performance of TRE in both prevention and response.

    The most difficult type of fraud to protect against is the one involving collusion between Seller and Account Debtor. In cases of collusion the verification/notification practices of the factor are of little preventative value in the short term. It's only when the first payment failures occur and are examined that the fraud is likely to surface.

    With respect to the specific cases you refer to, and the implicit question about TRE's practices and risks, I think that there are two issues to consider: what actually happened in these cases and what has TRE done in response.

    The issue of appropriate compensation for risk has to be INFORMED by the experience of the past but driven by the expectation for the future.

    If past practices might have increased certain types of risk, those are facts that cannot be ignored.

    If lessons learned have caused those practices to be changed, however, those new facts should also be acknowledged.

    Thanks very much for your comment.


  3. I received an anonymous comment that I felt contained a number of valid and important points but also included a couple of things that I felt were either inappropriate to make public under the terms of the Master Program Agreement or made reference to specific members of TRE management in a way that I felt appropriate only if the source was willing to be identified or if I could corroborate.

    So, I have edited the comment, leaving it as submitted with the exception of two deleted sections that are identified. CRL

    Anonymous has left a new comment on your post "Appropriate Compensation #1":

    I am always struck by your prose. Well stated. A couple of complementary points:

    1. Unfortunately, it is impossible to measure TRE's performance in terms of the prevention of fraud. The market will never know what it has prevented, only what has gotten through.

    2. While it is true that collusion is difficult to prevent, most factors rigorously establish debtor credit limits, and limit the impact of collusive fraud by those limits. The theory is that the stronger the debtor, the lower the likelihood of involvement in collusion. The weaker the debtor, the greater the likelihood of fraud (but the lower the credit exposure).

    Unfortunately, TRE buyers are unlikely to possess the same credit knowledge as a factor, and are far more likely to find themselves overexposed in a collusive scheme.

    3. It is difficult to know if TRE has learned any lessons because they do not want to publicize that they have experienced fraud. I refer you specifically to a recent article by Dyan Machan in Smart Money Magazine.

    Notice that when asked if they had experienced fraud they told Ms. Machan they had not.

    [I have omitted a portion of the balance of this comment, which I cannot verify. CRL]

    4. Risk management and credit management are not things I would imagine buyers would want TRE to acquire through experience, especially since it is the buyers who pay the price of inexperience.

    5. Fraud is but one risk. Operational mistakes due to growth, lack of archival knowledge and (once again) depth of experience, pose another very real and present risk.

    6. Unfortunately for buyers, TRE accepts no responsibility for the veracity of the information posted and traded on and through its platform. As a result, buyers will always be assuming far more risk than typical factors. The imbalance of risk and reward are skewed far too much to the detriment of the buyers and the benefit of sellers and TRE.

    [The balance of this comment is a lengthy quote from the TRE member documentation, which I believe is inappropriate to duplicate. Not that it isn’t important or on point. CRL]