Credit Risk and Machine Learning Concepts -8

Geoff Leigh
Analytics Vidhya
Published in
9 min readMar 13, 2020

--

How an Organization may leverage the Accounts Receivable position — Factoring and Credit facilities backed by A/R

ID 132340480 © Designer491 | Dreamstime.com Used under license.

As a business, the objective is to sell something at a profit to customers, repeatedly to the same target group and uniquely but to a large target group depending on the product or service. The income can be totally cash based and therefore reduce the risk, or with more normal trading terms for business-to-business that follows the ‘Quote — Order — Fulfill(make as necessary and deliver) — Invoice- Cash’ standard supply chain, with the creation of an open invoice and open accounts receivables between the fulfill/deliver when the item or service is likely to have a cost factor for you the supplier until the customer has paid the invoice and provided the cash for you. Then you can meet your obligations, such as payroll to your staff, payment to your suppliers for work in progress or items to stock, and any profits or dividends expected by your Shareholders and investors after paying any Taxes and Interest payments on loans.

So, you will always have a delay in payments to convert an order to cash, and the delay may have an impact to the amount of free cash that you can use to meet your obligations or continue to invest and expand your business. There are many ways of addressing this, other than insisting on pre-payment or the ability to draw a payment directly from the customer on delivery. These mechanisms are fine in a retail business to consumer setting, but not normal in a business-to-business setting.

So how does a business protect itself against bad debt? In my previous blogs in the series, I have stepped through the Credit Risk topic and the quantitative approaches that commonly are used to evaluate commercial risk, and introduced some newer topics in the field that include more stochastic consideration of probabilities of an event happening, and the Neural Network and Decision tree approach to including Qualitative as well as Quantitative information to determine the level of risk. As well as the level of risk of the speed and likelihood for a company that is a customer to be able to pay invoices promptly, the subset of these determinations include the prediction of a likelihood for the entity to cease operations and become a liability in terms of a bad debt when an open Accounts Receivables situation cannot be closed. So a few standard approaches on the Riskiness of your Account Receivables are the first step.

ID 34369674 © Kheng Ho Toh | Dreamstime.com Used under license.

How Risky are your Accounts Receivables?

The main metric to be evaluated is the Accounts Receivables Concentration factor. If the Bad Debt situation historically has been less than 1% of total book annually, then there is not a big problem to be aware of with existing customers and the collections processes that may already be in place, or the mitigation procedures such as factoring and A/R support of credit facilities that will be discussed in a little more detail further ahead.

The concentration ratio is calculated as the sum of the squares of each companies percentage of the overall Accounts Receivables balance. This number, expressed as a decimal, should be a value between zero and one, with a value closer to Zero indicating that the AR portfolio is less concentrated, which is a good indicator that the risk is spread so that one single customer is not a key source of income. A diversity index can also be derived, by dividing the concentration ratio by 1. The value returned will indicate that there are that number of key customers owning the most of the open Accounts receivables.

Taking an example, from an Article on American Express website by Mike Periu of Proximo LLC to help illuminate what this means:

Step 1: Calculate each company’s percentage of the overall AR balance

  • Company 1: $50,000 / $4,750,000 = 1.05%
  • Company 2: $50,000 / $4,750,000 = 1.05%
  • Company 3: $75,000 / $4,750,000 = 1.58%
  • Company 4: $100,000 / $4,750,000 = 2.11%
  • Company 5: $225,000 / $4,750,000 = 4.74%
  • Company 6: $230,000 / $4,750,000 = 4.84%
  • Company 7: $260,000 / $4,750,000 = 5.47%
  • Company 8: $260,000 / $4,750,000 = 5.47%
  • Company 9: $1,000,000 / $4,750,000 = 21.05%
  • Company 10: $2,500,000 / $4,750,000 = 52.63%

Step 2: Square each company’s percentage calculated in Step 1 and convert to decimals

  • Company 1: 1.05% * 1.05% = .011% = 0.00011
  • Company 2: 1.05% * 1.05% = .011% = 0.00011
  • Company 3: 1.58% * 1.58% = .025% = 0.00025
  • Company 4: 2.11% * 2.11% = .044% = 0.00044
  • Company 5: 4.74% * 4.74% = .224% = 0.00224
  • Company 6: 4.84% * 4.84% = .234% = 0.00234
  • Company 7: 5.47% * 5.47% = .3% = 0.003
  • Company 8: 5.47% * 5.47% = .3% = 0.003
  • Company 9: 21.05% * 21.05% = 4.432% = 0.04432
  • Company 10: 52.63% * 52.63% = 27.701% = 0.27701

Step 3: Add the squares obtained in Step 2

  • 0.00011 + 0.00011 + 0.00025 + 0.00044 + 0.00224 + 0.00234 + 0.003 + 0.003 + 0.04432 + 0.27701 = 0.33282

The accounts receivable concentration ratio in this example is 0.33282. The Diversity Index is 3.0046.

So, if the values of the AR Portfolio are of concern, the first strategy to consider is to sell to more customers, or create incentives to existing customers to pay faster by offering discounts for early payment. This assumes that the company has priced to consider the exposure and time to convert a sale to useful cash including sufficient margin to cover any borrowing or debt you may have to carry to enable profitability targets.

Essentially, by offering normal trade credit terms to customers, a company is in fact lending them money so that they may purchase goods or services from the company rather than from a competitor. I introduced the 3 approaches and the many subsets in my first blog in this series, to recap, there are 3 approaches commonly, and the problem of risk evaluation can be decomposed into 14 topics and sub-topics.

Approach 1: Entrust all or a subset of sales invoices to be covered by a commercial credit insurer or Finance Refactor of outstanding invoices

Approach 2: Use a commercial credit reporting agency to manage credit ratings and assume all risks for every sale transaction with mitigation by using Financial Services or Bank Revolving Credit or Business Loan guaranteed by AR Portfolio.

Approach 3: Use a combination of Commercial Credit Agency reporting, past trading history as an existing customer, non-structured external data with a mixture of rule-based calculations, mitigate exposure by using AR portfolio in Revolving Credit or Commercial Loan Guarantee

Commercial credit insurance takes a charge for each open account for an organization, and would give a maximum exposure credit limit and guarantee recovery of usually 80% of the open amount if the customer does not pay after a certain amount of time. Euler Hermes and AIG are the most prominent supplier of this facility, and the process involves submitting a customer for their approval and credit limit rating prior to coverage up to an agreed limit being provided by the insurance terms. This solution will eventually make a payment if the customer is unable or refuses to pay.

Accounts Receivable Factoring. This is a way of getting money for an open invoice at a discount, and either paying back the amount when the invoice is paid or letting the Factoring company own the debt and be responsible for any collection. Essentially that would mean discounting the total on every invoice to pay the factoring entity and taking only 80–90% of the cash on the invoice. Many major banks may offer this service to businesses, and there are even more non-bank financial services companies, some with dubious credentials but 1st Commercial Credit, Blue Vine, CIT Commercial Services are among some of the leading US Business Factor entities. The price that may be charged would depend on the business the Company is in, and the business and credit-worthiness of the customer whose invoices you wish to factor. There are two levels of factoring, Recourse factoring and non-recourse factoring, with recourse factoring usually cheaper in fees charged. When the factor bears all the risk of potential bad debts (this is non-recourse factoring), a higher fee or rate is charged to compensate for the risk. With recourse factoring, the company selling its receivables still has some liability to the factoring company if some of the receivables are not collectible, which could be because the product was faulty, the billing was incorrect or other dispute on the invoice was not resolved. The situation is commonly addressed in a start-up or small business where money is needed now, and the consequence of not having free cash would be more detrimental to the company.

(from Blue Vine’s website https://www.bluevine.com/).

Commercial Line of Credit: Major Banks such as Citi, Bank of America, HSBC or Wells Fargo have commercial loan products that they may make available to business clients. One category would be revolving credit, so that a business checking account does not necessarily have to have a cleared available balance to approve draws on the account that would otherwise be considered an overdraft situation, with the bank setting a maximum facility related to the information received on open Accounts receivables. A bank does this in the expectation that the debt position that they are allowing to exist will be temporary and is more than adequately covered by the expected income when the open AR positions are remitted in full by the customers. The company will have to pay bank fees and interest on the debt for the days that the account is overdrawn, but that will be a more favorable term than an unapproved overdraft situation might cost the company if they tried to make a payment when there was insufficient funds in the current accounts. Accounts Receivable, basically invoices that will be paid in 30 to 60 days, are seen by lenders as assets that can be used to secure financing. Lenders consider A/R a strong form of collateral because it is very liquid — it turns into cash quickly. To use your receivables as collateral, they must be high quality. Things such as the company procedures in opening credit accounts with customers including credit and other trade references, collections process as well as the full Quote-to-Cash system would have to be disclosed and approved by the Bank or Financial Services entity before providing this or the other type of commercial loan facility backed by Accounts Receivables.

Asset Based Loans: It may be that during the stages from new venture to mature successful companies one or both of the previous mechanisms may be used, rather than just Credit Insurance or taking on all credit risk. In between Factoring and Commercial Lines of Credit backed by A/R portfolio, is an Asset based loan. This may take into consideration the Accounts receivable, along with other assets such as inventory and machinery. An asset-based loan can behave like a line of credit if accounts receivable and inventory are used as collateral. Basically, you a loan account is set up that the business can draw from, as invoices and inventory become available. As with a factoring line, commonly only up to 80% of the value of the accounts receivable can be drawn, and it is expected to pay off the line plus applicable interest charges as cash becomes available. Asset — based loans are typically available for companies with more than $1 million in annual sales.

This is the 8th installment of my blogs on Credit Risk and Machine Learning. The next installment will consider how an entity is responding to changing markets and leveraging Artificial Intelligence.

The previous 7 installments may be found here:

https://medium.com/@geoff.leigh19/credit-risk-and-machine-learning-concepts-85ef47c978c7?source=friends_link&sk=5249acc679330bd64c76bcae1dc074d1

https://medium.com/@geoff.leigh19/credit-risk-and-machine-learning-concepts-2-fc37e1a05183?sk=94ef606e1c60e2cf1522b9c38a5e144e

https://medium.com/analytics-vidhya/credit-risk-and-machine-learning-concepts-3-d2bb2f39d843

https://medium.com/analytics-vidhya/credit-risk-and-machine-learning-concepts-4-3c44b479a3d1?source=friends_link&sk=cf6fe8b0a96d01c68971f72cbc179229

https://medium.com/analytics-vidhya/credit-risk-and-machine-learning-concepts-5-88f2dc1e18e2?source=friends_link&sk=2a4015bc86ee6071716865356ffb1a0d

https://medium.com/@geoff.leigh19/credit-risk-and-machine-learning-concepts-6-15adee7c0454?source=friends_link&sk=7f039a815c58ce5371c12ef5c72ac926

https://medium.com/analytics-vidhya/credit-risk-and-machine-learning-concepts-7-fdc9eb8dcd14?source=friends_link&sk=f3142319d5eb06540512264f587623e4

--

--

Geoff Leigh
Analytics Vidhya

Making Data into Actionable information and insight Over 30 years of Data and Systems engineering, development, consulting and implementation.