One of the most successful best practices for improving agency performance is the use of scorecards for assessing and rank ordering performance of agencies in competition with each other. Much like people, agencies thrive when they understand how they are evaluated, how to influence those factors that contribute to success, and the recognition and reward for top tier performance.
Rather than a simple view of performance based upon a recovery rate as a percentage of total inventory, best practice suggests that performance is more accurately reflected in vintage batch liquidation and peer group comparisons to the liquidation curve. Why? In a nutshell, differences in inventory aging and the liquidation curve.
Let’s explain this in greater detail.
Historically, collection agencies would provide their clients with better performance reporting than their clients had available to them. Clients would know how much business was placed in aggregate, but not by specific vintage relating to the month or year of placement. Thus, when a monthly remittance was received, the client would be incapable of understanding whether this month’s recoveries were from accounts placed last month, this year, or three years ago.
This made forecasting of future cash flows from recoveries difficult, in that you would have no insight into where the funds were coming from. We know that as a charged off debt ages, its future liquidation rate generally downward sloping (the exception is auto finance debt, as there is a delay between the time of charge-off and rehabilitation of the debtor, thus future flows are higher beyond the 12-24 month timeframe). How would you know how to predict future cash flows and liquidation rates without understanding the different vintages in the overall charged off population available for recovery?
This lack of visibility into liquidation performance created another issue. How do you compare the performance of two different agencies without understanding the age of the inventory and how it is liquidating?
An as example, let’s assume that Agency A has been handling your recovery placements for a few years, and has an inventory of $10,000,000 that spans 3+ years, of which $1,500,000 has been placed this year. We know from experience that placements from 3 years ago experienced their highest liquidation rate earlier in their lifecycle, and the remaining inventory from those early vintages are uncollectible or almost full liquidated. Agency A remits $130,000 this month, for a recovery rate of 1.3%. Agency B is a new agency just signed on this year, and has an inventory of $2,000,000 assigned to them. Agency B remits $150,000 this month, for a recovery rate of 7.5%.
So, you might assume that Agency B outperformed Agency A by a whopping 6.2%. Right? Er … no. Here’s why.
If we had better visibility of Agency A’s inventory, and from where their remittance of $130,000 was derived, we would have known that only a couple of small insignificant payments came from the older vintages of the $10,000,000 inventory, and that of the $130,000 remitted, over $120,000 came from current year inventory (the $1,500,000 in current year placements). Thus, when analyzed in context with a vintage batch liquidation basis, Agency A collected $120,000 against inventory placed in the current year, for a liquidation rate of 8.0%. The remaining remittance of $10,000 was derived from prior years’ inventory.
So, when we compare Agency A with current year placements inventory of $1,500,000 and a recovery rate against those placements of 8.0% ($120,000) versus Agency B, with current year placements inventory of $2,000,000 and a recovery rate of 7.5% ($150,000), it’s clear that Agency A outperformed Agency B.
This is why the vintage batch liquidation model is the clear-cut best practice for analysis and MI. By using a vintage batch liquidation model and analyzing performance against monthly batches, you can begin to interpret and define the liquidation curve.
A liquidation curve plots monthly liquidation rates against a specific vintage, usually by month, and typically looks like this:
Exhibit 1: Liquidation Curve Analysis
Note that in Exhibit 1, the monthly liquidation rate as a percentage of the total vintage batch inventory appears on the y-axis, and the month of funds received appears on the x-axis. Thus, for each of the three vintage batches, we can track the monthly liquidation rates for each batch from its initial placement throughout the recovery lifecycle. Future monthly cash flow for each discrete vintage can be forecasted based upon past performance, and then aggregated to create a future recovery projection.
The most sophisticated and up to date collections technology platforms, including Experian’s Tallyman™ and Tallyman Agency Management™ solutions provide vintage batch or laddered reporting. These reports can then be used to create scorecards for comparing and weighing performance results of competing agencies for market share competition and performance management.
As we develop an understanding of liquidation rates using the vintage batch liquidation curve example, we see the obvious opportunity to reward performance based upon targeted liquidation performance in time series from initial placement batch.
Agencies have different strategies for managing client placements and balancing clients’ liquidation goals with agency profitability. The more aggressive the collections process aimed at creating cash flow, the greater the costs. Agencies understand the concept of unit yield and profitability; they seek to maximize the collection result at the lowest possible cost to create profitability.
Thus, agencies will “job slope” clients’ projects to ensure that as the collectability of the placement is lower (driven by balance size, customer credit score, date of last payment, phone number availability, type of receivable, etc.) For utility companies and other credit grantors with smaller balance receivables, this presents a greater problem, as smaller balances create smaller unit yield. Job sloping involves reducing the frequency of collection efforts, employing lower cost collectors to perform some of the collection efforts, and where applicable, engaging offshore resources at lower cost to perform collection efforts.
You can often see the impact of various collection strategies by comparing agency performance in monthly intervals from batch placement. Again, using a vintage batch placement analysis, we track performance of monthly batch placements assigned to competing agencies. We compare the liquidation results on these specific batches in monthly intervals, up until the receivables are recalled.
Typical patterns emerge from this analysis that inform you of the collection strategy differences. Let’s look at an example of differences across agencies and how these strategy differences can have an impact on liquidation:
As we examine the results across both the first and second 30-day phases, we are likely to find that Agency Y performed the highest of the three agencies, with the highest collection costs and its impact on profitability. Their collection effort was the most uniform over the two 30-day segments, using the dialer at 3-day intervals in the first 30-day segment, and then using a balance segmentation scheme to differentiate treatment at 2-day or 4-day intervals throughout the second 30-day phase. Their liquidation results would be the strongest in that liquidation rates would be sustained into the second 30-day interval.
Agency X would likely come in third place in the first 30-day phase, due to a 14-day delay strategy followed by two outbound dialer calls at 5-day intervals. They would have a better performance in the second 30-day phase due to the tighter 4-day intervals for dialing, likely moving into second place in that phase, albeit at higher collection costs for them.
Agency Z would come out of the gates in the first 30-day phase in first place, due to an aggressive daily dialing strategy, and their takeoff and early liquidation rate would seem to suggest top tier performance. However, in the second 30-day phase, their liquidation rate would fall off significantly due to the use of a less expensive IVR strategy, negating the gains from the first phase, and potentially reducing their over position over the two 30-day segments versus their peers.
The point is that with a vintage batch liquidation analysis, we can isolate performance of a specific placement across multiple phases / months of collection efforts, without having that performance insight obscured by new business blended into the analysis. Had we used the more traditional current month remittance over inventory value, Agency Z might be put into a more favorable light, as each month, they collect new paper aggressively and generate strong liquidation results competitively, but then virtually stop collecting against non-responders, thus “creaming” the paper in the first phase and leaving a lot on the table.
That said, how do we ensure that an Agency Z is not rewarded with market share? Using the vintage batch liquidation analysis, we develop a scorecard that weights the placement across the entire placement batch lifecycle, and summarizes points in each 30-day phase.
To read Jeff's related posts on the topic of agency management, check out: