As I discuss in this paper, international engagement has served as a countervailing force against these. Because of this, it is very likely that elections in PNG would be worse still without Australian involvement.

This doesn’t mean that Australian efforts can’t be improved. Here are some suggestions. Because this is a blog post they are, by necessity, broad. I hope many others will offer a lot more in coming months; for now, treat these ideas as a start.

First up, recognise the road to the next elections starts today. Improving them will be an ongoing effort. It will require engagement, pressure and assistance, every year from now to 2022.

The chart above shows my estimates of Australian aid spending devoted to elections in PNG since 2002. As you can see, the post-2012 effort was inconsistent and less than previous elections.

This isn’t the aid program’s fault. In between aid cuts and the death of AusAID, it was a tumultuous time.

Even so, there’s still a lesson from 2017: improving elections requires substantial, sustained engagement. It requires staff devoted to the task and it requires the steady accumulation of contextual knowledge.

Speaking of knowledge, it’s crucial to find out exactly what went wrong in 2017. I’ve already outlined the general dynamic that I think drives PNG’s electoral issues. This, I think, explains a lot. Nevertheless there is more to be learnt about the specific nature of the specific woes of 2017.

Some people have alleged that there was major, centralised subversion of electoral system coordinated by senior politicians. I’m inclined to think the real problems were primarily neglect and localised capture, possibly with centralised assistance.

I could be wrong though. I want to know exactly what happened. And the Australian government aid program definitely needs to know. This knowledge will be crucial in shaping future attempts to improve things.

Different issues will have different remedies. Australia won’t be able to provide all of them; some will come from within PNG. But getting international engagement right means knowing exactly what went wrong.

I’ve met many DFAT staff who understand PNG’s politics well. What I’m talking about goes beyond this, though. I think the real dividends will come from a careful and comprehensive synthesis that transcends what individuals know, and which will be a source of understanding long after specific people have moved to new roles.

Then, one way or another — in the wake of the 2017 mess — some form of political pressure from Australia will also be necessary. I understand it has to be gentle, I understand there’s an ongoing constructive relationship to be maintained. But some of the needed electoral improvements will only come if PNG’s politicians are pressured.

Ideally, this pressure will come from within PNG, and I’m sure some of it will. But given the nature of voter-politician relationships that I spelled out in my first blog, pressure from within is unlikely to be enough on its own.

International pressure will also be needed. Maybe this doesn’t fit particularly well with Australia’s other interests in PNG. But — trust me — Australia’s nearest neighbour becoming a failed democracy will be worse again still for Australia. Much worse.

Pressing PNG’s political actors to improve the quality of the country’s democracy might be awkward, but it is essential. For Australia as well as PNG.

Commit, learn, and pressure. Not rocket science. But all essential aspects of Australia helping Papua New Guinea on the road between now and much better elections in 2022.

PNG itself is not short of talented and dedicated people who will be at the forefront in bringing better elections in five years’ time. But it also has a deeply dysfunctional political economy.

An engaged, knowledgeable, well-intended neighbour, who is willing to lean on political elites when required, will help domestic reformers a lot.