It is unlikely Labor's climate change policy was the main reason why, against all expectations, it failed to win government. A detailed analysis of voters by the Australian National University released last month found views on the leaders and candidates were more important. Similarly, the Australian Election Study reveals a long term trend in voter unhappiness with the major parties: leading to greater volatility and uncertainty in election outcomes.

Nevertheless, elections are not just popularity contests between leaders. Policies matter; Labor is reviewing them. Last week Opposition Leader Anthony Albanese said "... what you need to do is to re-examine all of your policies...we'll take a strong position on climate change ... by definition the specifics in 2022 of going forward are different from what it was in 2015 when our policy was announced."

Tony Abbott, in his concession speech (after losing to an independent who campaigned strongly on climate) said "where climate change is a moral issue, Liberals do it tough. Where climate change is an economic issue we do very well." Some commentators have suggested Labor lost ground among workers affected by the economics of climate policies. More recent evidence does not support this. Opinion polls suggest concern about climate change in Australia is widespread, including among working class voters, and growing.

Here's a different possibility: voters worried by global warming want climate change policies - just not those advanced by Labor.

Labor's policies were not easy to understand. The "Climate Change Action Plan" had an Emissions Trading Scheme as its centrepiece (see page 12 of the action plan's fact sheet). It was complicated, with trades, caps, offsets and other mechanisms. Labor promised further work on rules "governing the allocation of caps to liable entities, access to international markets, the operation of the domestic offsets market and the like".

Labor has been attracted to complicated emissions trading schemes for a while. It proposed the Carbon Pollution Reduction Scheme under the Rudd government, in a 2008 White Paper, but ultimately failed to win support for the scheme. The alternative carbon pricing scheme introduced by Julia Gillard in 2012 was short lived, abolished by the Coalition government in 2014. The CPRS was a market-based mechanism for emissions reductions, allowing trading of permits

Labor's own minister for the CPRS struggled to explain its details. Designed by technically proficient economists, delighted by the prospect of creating a market to drive better and more innovative solutions, it lost its way when exposed to public debate. It's not hard to see why. The messages are mixed. If carbon emissions are bad, then why should a government give people permits to do something bad? We do not take the same approach with other types of pollution, or indeed other activities society considers undesirable. We don't for example give citizens a tradeable permit for a little bit of violence. In a violence trading market, peaceful people could sell their permits, and rich people could buy enough permits to get away with murder (literally). By gradually reducing the cap on permits each year we would bring down the murder rate. Lovely in theory, implausible in practice.

To take a real world example, the Australian government's very successful campaign to discourage smoking did not set up a market in smoking permits; rather, it applied a combination of information campaigns and taxes. Ultimately, for many people trade in carbon is morally offensive. In theory a carbon tax and a trading system are equivalent in their effect: the first sets a price on, the second controls amounts of, emissions. They both price carbon. In practice, they feel different to voters.

Australians are wise to problems of market design. Markets can be highly effective, encouraging innovation and driving low cost solutions. In environment policy, taxes, charges, even container deposit schemes are all examples of market-based mechanisms. Often they work. However, the larger the markets and the more complicated the market-based mechanisms, the more likely they will be to fail.

Marketisation of the vocational education and training sector, for example, was a policy pursued for laudable reasons: encourage new providers, new curriculum, better teaching methods and system innovation. In some cases it worked, encouraging excellent, innovative and creative new vocational education providers. Sadly, other providers emerged which were run by shady characters seeking to milk the system and defraud students. As a result, many students have been left with useless qualifications and high bills, governments left with bad debts and bailouts. A basic principle of policy design was forgotten: when government puts a pile of money on the table it needs rigorous rules governing access to that money, and systems of accountability to ensure results are achieved. Governments need not only to set those rules but monitor and enforce them.

A similar example can be found in the national electricity market. Its creation now more than 20 years ago was based on the premise that competition, especially among electricity generators, would drive down costs and encourage better practice. It separated generation, transmission and sale of electricity into three separate streams of business, with different markets.

It aimed to replace inefficient state based monopoly providers and support an interconnected east cost grid. Initially working as intended, it has been showing signs of strain: criticised by consumers, failing to deal with renewables, overly complex, with electricity regulators subject to frequent legal challenges. Litigants observe, probably in jest, that lawyers generate more profits for energy companies than engineers. Partly this is because of energy policy confusion under successive governments, both Coalition and Labor; partly the impact of constantly changing climate policies on the energy market; part a legacy of federal rivalries. One source of complexity has been ministers and legislatures at Commonwealth and state level attempting to resolve problems with regulation by patching over new regulation: losing sight of the scheme's original objectives.

Every time a new and complex government scheme is rolled out these design issues need to be considered. The National Disability Insurance Scheme was set up with the best of intentions, but problems are emerging. Quite likely the same kind of fraudsters who sought to profit from the vocational training market are eyeing off the NDIS. The NDIS Quality and Safeguards Commission, an independent statutory body set up to oversee the quality and safety of services, will need to be both vigilant and tough. The management of the National Disability Insurance Agency - criticised recently for excessive spending - needs close oversight to guard against it becoming captured by providers. That is always a danger when the service providers have far deeper pockets, and can spend big on lobbying, than the clients.

Technocrats, public servants with deep technical expertise (such as economists, engineers or scientists) are often attracted to highly complicated programs. They are an intellectual challenge and have high theoretical content. This kind of program more often than not fails in the real world. It is a lesson Labor should have learned from the CPRS. It is clear from the Prime Minister's speech to the public service on August 19 that he too is aware of the dangers of a theoretical over a practical solution. He said "the ultimate test of a strategy is not how pretty it looks, but how well it's done". He could have observed, but for obvious political reasons did not, that ministers too should heed that advice. One reason why large and overly ambitious programs are launched with little thought about design principles is that ministers want them, driven by a relentless media cycle and demands for "announceables".

Sadly, with some ministers the bigger the program the better, they see it as success. This is a mistake. There are two important design principles ministers would do well to learn: small and simple. A program is more likely to work if it is small and targeted, and as simple as possible. This does not mean simplistic - sometimes programs have to be complex to work. If a program needs only five elements, well and good; if 5000, including independent oversight and review, is the genuine minimum required, also good. The basic principle is to keep design as simple as is necessary to achieve a program's objectives. If the Prime Minister applies this we will all be better off.