Global menu

Our global pages


The dangers of salary benchmarking: Hong Kong Competition Commission advises HR trade associations not to publish salary projections

  • Asia
  • Competition, EU and Trade



It has recently been reported that the Hong Kong Institute of Human Resource Management (HKIHRM) and the Centre for Human Resources Strategy and Development both received advice from the Competition Commission earlier this year that publication of salary increase forecasts may be risky from a competition law perspective. Although the Competition Commission have not commented publically, it is apparent that it is aware of the potential competition law issues that arise in the context of salary benchmarking as discussed in our Competition Law Issues in HR Seminar we ran in April 2016.


The Competition Ordinance came into effect in December 2015 and introduced cross-sector competition law rules in Hong Kong prohibiting:

  • agreements between businesses which have the object or effect of preventing, restricting or distorting competition in Hong Kong (“the First Conduct Rule”); and
  • companies with market power, including monopolists, abusing their power (“the Second Conduct Rule”).

There are very severe penalties for breach of either conduct rule, including financial penalties of up to 10% of Hong Kong annual revenues for each year of infringement - up to a maximum of three years.

Whilst at first glance, the exchange of salary and bonus information may result in reduced costs for the company (and consequently the consumer), caution should still be exercised with this type of information exchange. If labour is viewed as a ‘market’, then it follows that employees are assets to be ‘bought’ at a price (i.e. their salary and bonus). Actions which could artificially depress the price paid by employers for staff could (at least theoretically) be anti-competitive and a potential breach of the First Conduct Rule.

Publication of forecasts

According to recent press reports, both trade associations mentioned above conducted market salary surveys in Hong Kong and used the data to publish an annual indicative average percentage pay rise. The Commission’s firm message is that trends should be linked to historical data only and not communicated as future forecasts.

These reports strongly support the view that the Competition Commission considers breaches within the HR sphere to be within its remit.

Reminder of practical tips

As a result of this risk, we recommend employers follow the following guidelines to ensure that they stay on the right side of the line as far as sharing salary data is concerned.

Don’t: agree salary or benefit levels with competitors

Agreeing salary or benefit levels with competitors is effectively price-fixing which is serious anti-competitive conduct under the Competition Ordinance. Remember that an ‘agreement’ does not have to be written down, a ‘meeting of minds’ will suffice. This could be an unspoken expectation or a course of conduct.

Do: use an independent third party

Data on employee salaries and benefits should not be exchanged directly with other employers. If you are asked by a competitor to share information – you should politely refuse! Using an independent third party that pledges not to disclose one company’s strategically useful information to the other, reduces the risk.

Do: use historical information

Even when using an independent third party, data exchanged as part of a benchmarking exercise should be “historic”. Exchanging information on prospective or planned salaries and benefits is risky. What is considered "historic" may differ according to particular industry practice. However, information over 12 months is generally considered to be historic as most employers operate annual pay reviews.

Do: ensure that data is not attributable

Even where an independent third party is conducting the benchmarking survey, it is vital that information cannot be reverse engineered to identify its source.

This can be achieved by:

  • having at least five participants to the survey so that information is sufficiently diluted;
  • anonymising results so that it cannot be tracked back to individual employers; and
  • presenting data in an aggregated format in ranges or averages.