Lextree Simply the best software for managing legal entities and documents.
The Berkman Letter Practical analysis and commentary for business, law, and more.

What is the Altman Z-score, and how well does it predict the likelihood of bankruptcy?


I'm evaluating the strength of the tenant in a building I want to buy.


The Altman Z-score is an oft-used formula that aggregates balance sheet data of a company to predict the likelihood of bankruptcy in 1 or 2 years. The accuracy of the Altman-Z score is generally better at the 1 year mark than the 2 year mark. The Z score has been revised a few times to address two short-comings of the original version. First, the original data focused on public companies so it was adjusted to cover private companies. Second, the score was adjusted to apply to companies outside of manufacturing since that was the origin focus.

Academics and creditors have validated the Z score for predicting bankruptcy. Depending on the company and your concerns, you might consider a few points:

  1. The Z score depends heavily on the quality of an individual company's financial data. If they can't give you 3 years of audited financial statements, then the Z score might not give you what you want.
  2. For small companies (maybe revenue < $5-10M), the Z Score is generally considered unhelpful, because the financial condition of the owner or principals is far more important. Most creditors and bankers will use a combination of personal tax returns and Fair Isaac Small Business Scoring Service (or competitive solution).

For small companies , the tax returns help establish the credit worthiness of the individual. The SBSS score uses aggregated credit data (much like a personal credit score) to build a predictive, statistical model.

All of these scores, Z, SBSS, and the like focus on Probability of Default (or bankruptcy in the case of the Z score). They do not address Loss Given Default (LGD) or Exposure at Default (EAD). These two values may not be helpful in a tenant situation but the do help put the default in context. In order words, default by the tenant in a single unit building is very different from that same tenant defaulting in a large, multi-tenant building, because the rental income does not go to zero in the later case.

I've included links to a helpful description, a paper by Altman on the score, and to the FICO SBSS score information.


The Altman Z-Score is a model for predicting corporate bankruptcies based primarily on the quality of the balance sheet. The Altman Z-Score has been shown to be successfully predictive of bankruptcy about 75% of the time. While the Altman Z-Score is the most widely used model of its kind, there are other similar models, some of which have shown better performance than the Altman Z-Score, such as the CHS model, the Ohlson O-Score, and Merton's Distance to Default. However, many of these alternatives are more complex than the Altman Z-Score and incorporate metrics that are only available for publicly traded companies.