With the federal election now set for August 21 the AMP's chief economist, Dr Shane Oliver says it is natural to wonder what impact, if any, there might be on investment markets.

The historical record suggests little clear impact, with other factors proving more important. This may partly reflect the broad similarity in terms of macro economic policies between both sides of politics.

History provides little reason to think the outcome of the election later this year will have a lasting impact on the share market.

Over the post war period the average return for the Australian share market under the Coalition has been 13.2% pa, compared to 9.9% pa under Labor governments.

However, if the current term is excluded on the grounds it has been dominated by the global financial crisis - which by definition has been global - then the average under Labor rises to 12.6%.

In other words there is not much difference between the two major parties.

Conventional perceptions that conservative governments are always better for shares don't always play out.

While the Hawke/Keating period of 1983 to 1996 saw very strong share market gains, with an average return of +17.1% pa, the Fraser Liberal/National government of 1975 to 1983 saw shares return just 2.2% pa.

Of course, the global backdrop was poor during the Fraser years and good during the Hawke/Keating period so other factors tend to dominate the impact of domestic politics.

Moreover, once in government political parties of either persuasion are usually forced to adopt sensible macro economic policies if they wish to deliver rising living standards to their constituents.

Both the Coalition and Labor agree on the key macro fundamentals - i.e. the need to keep inflation down, to return the budget to surplus and in the benefit of free markets.

The key policy impact of a change of government in Australia is more likely to be felt at the sector level than at the macro level.

But what about the uncertainty created by the election itself?

Of course elections have the potential to have a short-term impact on investment markets.

This stems mainly from the fact investors don't like the uncertainty associated with the prospect of a change in government.

The next chart shows Australian share prices from one year before till six months after Federal elections since 1983.

This is shown as an average for all elections (but ex 1987 and 2007 given the global share crash 3 months after 1987 the election and the start of the global financial crisis in 2007), and the periods around the 1983 and 2007 elections, which saw a change of government to Labor, and the 1996 election, which saw a change of government to the Coalition.

There is some evidence of a period of flat lining, possibly reflecting investor uncertainty, in the run up to elections followed by a relief rally soon after it is over.

However, the elections when there has been a change of government have seen a mixed picture.

Shares rose sharply after the 1983 Labor victory but fell sharply after the 2007 Labor victory, with global developments playing a big roll in both.

After the 1996 Coalition victory shares were flat to down.

The point is it's not obvious which party is best for shares, and in any case historically the impact of swings in global share markets arguably played a much bigger role.

The next chart shows the same analysis for the Australian dollar.

In the six months or so prior to Federal elections there is some evidence the $A experiences a period of softness and choppiness which is consistent with uncertainty about the policy outlook, but the magnitude of change is small - just a few percent.

In any case, on average, the $A has drifted sideways after elections.

While the $A fell soon after the 1983 Labor victory this was due to a policy devaluation in the dying days of the fixed exchange rate system.

The next chart shows the same analysis for Australian bond yields. Interestingly, on average bond yields have drifted down over the six months prior to Federal elections since 1983.

The average decline has been around 0.75% which is contrary to what one might expect if there was investor uncertainty regarding the policy outlook.

However, the tendency for bond yields to decline ahead of Federal elections appears to be more related to the aftermath of recessions, growth slowdowns and/or falling inflation prior to the 1983, 1984, 1987 and 1993 elections and the secular decline in bond yields through the 1980s and 1990s in general.

More broadly, it's hard to discern any reliable affect on bond yields from Federal elections.

Policy differences and sectoral impacts

While it is doubtful the election will result in a significant impact on financial markets at an aggregate level, there is more potential for a significant sectoral impact.

The main areas of difference between the two parties are likely to be in terms of the health, the mining tax, the increase in the superannuation guarantee from 9% to 12%, the national broadband network, policies on climate change and paid maternity leave.

A Coalition government may be more relaxed in terms of industry regulation but may take a more measured approach to stimulating the economy should there be another global economic downturn.

In terms of the aggregate impact on business and the economy, these policy differences are likely to largely offset each other reinforcing the assessment that the outcome of the election is unlikely to have a significant impact on the economy or financial markets at an aggregate level.

This may not have been the case if the Labor Government had not pushed out the ETS, heavily watered down the mining tax and wound back the "Big Australia" concept.

Against this though, the policy differences do suggest the potential for significant sectoral impacts.

Conclusion

Short of the adoption of more significant policy differences from either side of politics it's hard to see the election having a big impact at a macro level, beyond individual sectoral impacts.