Syntha Lab RU

Supplier concentration risk (HHI) and expected loss

From the suppliers’ shares of the buy counts the Herfindahl–Hirschman index (HHI) and the effective number of suppliers, the expected margin loss if the largest fails (probability × share × margin × share of the season without supply) and compares it with the cost of a second source: the price surcharge on the moved volume.

Choosing a factory and quality For whom: Procurement and risk directors.

Calculate on your data →

At a glance

What question does it answer?
How concentrated the buy is, what a failure of the largest costs and whether a second source pays off.
What data is needed?
Supplier shares.
How to read the result?
HHI and the effective number of suppliers, then the loss against the cost of a second source. Example on the default values: HHI: 0.3; Effective number of suppliers: 3.33; Share of the largest: 45%; Expected margin loss: RUB 553,846. The figures are illustrative: enter your own to see your result.

Why and how to reason

The share of the largest supplier is not everything: what matters is what a failure costs and how long a replacement takes. In the example HHI is 0.300, the effective number of suppliers 3.33, the expected loss 0.55 m RUB. A second source pays off if it costs less than this loss.

When to use it

Apply when revising the supplier base and before launching a new collection at one factory.

How to read the result

Look at HHI and the effective number, then the loss against the cost of a second source and the break-even surcharge.

Common mistakes

They count risk by share only, not by the switching time, and treat a second source as free.

What data is needed?

  • Supplier shares.
  • Probability and duration of failure, margin.
  • The second source surcharge.

Inputs

FieldUnitValue in the exampleMeaning
Supplier 1%45Share of the buy by value. The shares are rescaled to 100%; an empty field is zero. The example value is illustrative.
Supplier 2%25Share of the buy by value. The shares are rescaled to 100%; an empty field is zero. The example value is illustrative.
Supplier 3%15Share of the buy by value. The shares are rescaled to 100%; an empty field is zero. The example value is illustrative.
Supplier 4%10Share of the buy by value. The shares are rescaled to 100%; an empty field is zero. The example value is illustrative.
Supplier 5%5Share of the buy by value. The shares are rescaled to 100%; an empty field is zero. The example value is illustrative.
Supplier 6%0Share of the buy by value. The shares are rescaled to 100%; an empty field is zero. The example value is illustrative.
Probability of failure per season%10Your estimate: a missed delivery, sanctions, a closed port. There is no single norm. The example value is illustrative.
Weeks without supplyweeks8How many weeks switching to a replacement takes: count the time, not only the share. The example value is illustrative.
Weeks in the seasonweeks26The length of the period the margin refers to. The example value is illustrative.
Margin for the period₽40,000,000Gross margin of goods that depend on this source over the season or the year. The example value is illustrative.
Purchases from the largest for the period₽120,000,000For the surcharge on the moved volume. If all purchases are 120 m and the largest has 45%, enter 54 m. The example value is illustrative.
Part of the largest supplier’s volume you move%50The share of volume you keep at a second source permanently. The example value is illustrative.
Second source surcharge on the price%3How much more expensive than the largest, including qualification and logistics. The example value is illustrative.

How it is calculated

StepFormula
HHIthe sum of squared supplier shares; effective number = 1 ÷ HHI
Expected lossprobability of failure × share of the largest × margin × (weeks of failure ÷ weeks of the season)
Second sourcecost = surcharge × purchases from the largest × share moved; benefit = loss × share moved
Decisiondiversify if the benefit exceeds the cost

Worked numbers

Default values from the app; in the interactive version you enter your own.

HHI
0.3The sum of squared shares (as fractions of one).
Effective number of suppliers
3.331 ÷ HHI: that many equal suppliers would give the same concentration.
Share of the largest
45 %
Expected margin loss
553,846 ₽Probability × share × margin × (weeks of failure ÷ weeks of the season).
Loss removed by a second source
276,923 ₽
Cost of a second source
810,000 ₽Surcharge × moved volume.
Decision
does not pay off yet
Break-even surcharge
1.03 %
HHI after the move
0.199
Supplier shares, %
Bar chart: 1 45, 2 25, 3 15, 4 10, 5 5, 6 00102030405011Share: 454522Share: 252533Share: 151544Share: 101055Share: 5566Share: 00%
Data as a table
MeasureShare
145
225
315
410
55
60

Suppliers’ contribution to HHI

SupplierShareSquareContribution to HHI
Supplier 145 %0.202567.5 %
Supplier 225 %0.062520.8 %
Supplier 315 %0.02257.5 %
Supplier 410 %0.013.3 %
Supplier 55 %0.00250.8 %
Supplier 60 %00 %

At this surcharge a second source costs more than the expected loss (810,000 ₽ vs 276,923 ₽): it does not pay off; it would pay off at a surcharge of no more than 1.03 %.

The model treats the second source as independent: it does not fail together with the largest and manages to take the moved volume. A rare but heavy failure and insurance cases are poorly shown by the expected loss: look at the worst case too.

What to do with the result: if HHI is high and switching takes more than a month, qualify a second supplier in advance for a small volume. A typical mistake: counting risk only by share, not by the replacement time.

Nuances and limits

  • Counting risk by share only.
  • Ignoring the switching time.
  • Treating a second source as free.

Terms used

Calculate on your data →