Purchase Edition

Edition 56

Contents
Online Only

An algorithm for altruism

IN 2013, FORMER US President Barack Obama identified inequality as the defining challenge of our time, and claimed it as a cornerstone motivator for his administration. Without the actions taken by the Obama administration during its two terms to reign in the acceleration of economic inequality in the US and abroad, circumstances would no doubt be more dire than they are now – though inequality remains on the rise. The role of this trend has been widely commented on as the basis for the global political upheaval that began in 2016.

In 2015, Roger Wilkins of the Melbourne Institute of Applied Economic and Social Research observed that economic inequality has been increasing faster in the US and the UK over the last decade than it has in Australia; nevertheless, in Australia it is currently the highest it’s been in the postwar period. Where traditionally the income gap has been small in comparison to other countries, in the last decade it has widened at an alarming rate.

In a special event held at the Sydney Opera House in 2016, the French economist and author of the acclaimed Capital in the Twenty-First Century (Harvard University Press, 2014) Thomas Piketty noted that income inequality in Australia reached its highest level since 1951 when, in 2013, the total income of the top 10 per cent of Australians was equal to half of the total income of all Australians. He called for public policy to be geared towards addressing the widening inequality problem. Outraged Sydneysiders jumped up and cheered: their emerald city having become emblematic of wealth inequality in recent years. As home to both the most socio-economically advantaged (Ku-ring-gai) and disadvantaged (Fairfield) urban local government areas in NSW, inequality in Sydney has been worsening at a pace that even the millennial generation has been able to witness.

Widening inequality in Australia has been accompanied by twenty-five years of overall economic growth, which critics of Piketty say has in turn helped drive trickle-down prosperity. That may be so, but the data shows the trickle is rather more like a drip. At a glance, that top 10 per cent of earners took home only 24 per cent of the national total in 1980 (their proportional share has been rising by just under 2 per cent per year since then). Contrast that with the findings of Deloitte Access Economics in successive reports prepared for Food Bank that the number of Australians accessing food relief services has been on the rise since before 2012, not just in absolute numbers but also as a proportion of the total population of Australia.

Research from Homelessness Australia and the Australian Bureau of Statistics found homelessness increased in New South Wales and Victoria by a fifth over the last decade, there are almost three million Australians living in poverty, and 17.4 per cent of all Australian children aged under fifteen are living below the poverty line.

 

A QUICK SCAN of the data can help us see what’s down the line. In its 2016 research paper, The Role of Capital and Labour in Driving Economic Growth in Australia, professional services firm KPMG contends that without a turnaround in Australia’s falling productivity growth, and with no new mining or resource booms on the visible horizon, living standards of Australians are at serious risk of decline. Australia’s ageing population, they point out, is among the most significant exacerbating factors. In 2016, the McKinsey Global Institute’s report Poorer Than Their Parents? Flat or Falling Incomes in Advanced Economies claimed that millennials will likely be the first generation in seventy years to enjoy less prosperity than their parents.

The intergenerational wealth distribution is skewed towards older generations. The Grattan Institute, in its 2014 report The Wealth of Generations, found that ‘households aged between sixty-five and seventy-four today are $200,000 wealthier than households of that age eight years ago. Meanwhile, the wealth of households aged twenty-five to thirty-four has gone backwards.’

Compounding this is the fact that, according to the Graduate Destinations Report 2014 from Graduate Careers Australia, full-time employment rates for university graduates four months after graduation were the lowest on record since their survey began in 1982. The previous low immediately followed the recession of the late ’90s. ‘Overqualified, under-utilised and poorly paid,’ as ABC’s business editor Ian Verrender describes younger skilled workers and new graduates, lamenting recent rises in part-time work and ‘underemployment’ as symptomatic of an ongoing employment uncertainty.

Characterising the job market millennials face today as weak is a bit of an understatement. But it isn’t all downward trends: according to the 2016 BNP Paribas Global Entrepreneur Report, millennial entrepreneurs – ‘millennipreneurs’ – are on average, compared to baby boomer entrepreneurs, starting their first business around eight years younger, starting twice as many businesses, and employing four times as many people. Imagine what a millennial-driven economy could achieve in Australia with the endorsement of investment availability.

 

DIGITAL, DATA-DRIVEN AND machine-learning technologies native to the millennial experience, and core to ‘millennipreneur-ism’ experimentation, are also significantly accelerating the inequality gap – and the impact of this is yet to be fully felt.

A growing panoply of profit-motivated products and services, powered by big data and machine-learning algorithms, now saturates the mainstream. High-frequency trading is a prime example: an automated trading technique used by banks and institutional investors to analyse enormous volumes of data from global markets and make millions of trades in seconds or less. A whole ecosystem of trading algorithms has evolved to influence, disrupt and prey on human traders and on one another with terrifying speed, giving the wealthiest players in the global financial system – those who can afford the technology– an enormous advantage in the ‘open’ market. By anticipating and exploiting identified trends, using what’s known as ‘predictive analytics’ and ‘prescriptive analytics’, those well-heeled, well-resourced, top-of-the-pyramid private banks and financial institutions and their clients (many of whom are the super rich) make enormous returns on portfolios of often obscure and complex financial products.

Naturally enough, finance was among the first industries to embrace machine learning algorithms in the commercial sector. Today, the use of these tools is in hyperdrive, accelerating into new industry areas like ‘fintech’ or ‘insurtech’ (innovations in finance and insurance), online and offline retail environments, digital marketing, social media and consumer brands. Data is now widely valued as an asset class of its own. Various more-or-less traditional forms like operational data, customer and client data, research findings and web traffic are suddenly a coveted and readily tradeable commodity. Add to the mix the Internet of Things rapidly assuming control of the built environment around us and we have a world of interconnected complexity beyond the everyday imagination.

This technological surge is radically changing the human element of tomorrow’s workforce. Ray Kurzweil, director of engineering and chief futurist at Google, expects robots to achieve human-level intelligence by 2029. Global IT research and advisory company Gartner predicts that this will have displaced up to one third of jobs by 2025. Andrew McAfee and Erik Brynjolfsson, co-founders of the MIT Initiative on the Digital Economy and authors of The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies (WW Norton & Company, 2014), have called for new policies to protect the vulnerable. They argue that low-wage jobs are especially at risk. Countries that are prolific producers and consumers of technology, they say, will feel the effects first. Even supposing that these new technologies create some jobs to replace those they displace, this is still likely to result in an overall increase in the proportion of the population dependent on welfare of one kind or another.

In Australia, according to the Baseline Valuation Report prepared by PwC for the Department of Social Services, millennials have inherited a social security system whose beneficiaries as at 2015 were estimated to represent a $4.8 trillion cost over the course of their lives to Australian taxpayers. The same report offered, by way of scale, the estimated Australian GDP in 2015 of only $1.6 trillion. It’s a sobering thought. How will it be possible to maintain and improve the wellbeing of all when there is such a heavy burden to carry, and the data indicates the situation is worsening? Something must be done.

 

AUSTRALIA’S SOCIAL SERVICES sector, including those government services and not-for-profits delivering community services, has not kept pace with the digital and data revolution. To be fair, neither have many profit-motivated industries – but the challenges are greater for the social sector because of systemic barriers to entry, foremost of which are technical skills and funding.

Traditionally, the social services sector has not invested in digital and data solutions due to limited funding. The Australian Charities and Not-for-profits Commission estimates that the sector receives around 83 per cent of its total funding ($25.5 of $30.6 billion annually) through direct government grants. Corporate and philanthropic foundations are often the alternative funders that enable not-for-profits and service providers to build technical skills and capacity in the domain of data science and analytics. A small number have embarked on this journey, but it is early days.

Only recently have governments begun to see the possibility of using data and analytics for improved service delivery; previously they had mostly been limited to investing in data-sharing across government agencies. The NSW Department of Finance and Services launched the Data Analytics Centre, which facilitates data-sharing between agencies to inform more efficient, strategic and whole-of-government evidence based decision-making. This has yet to be extended to the wider social services sector.

Government policy and programs have long been criticised as ‘bandaid’ solutions, delivered on the basis of inaccurate or unavailable data. Short-term political cycles deliver policies and programs from on high that are not designed around the needs of any individual community, and are chopped and changed before real impact can be felt and measured. Nor do these policies consider the deep expertise of the people in the community and the evidence they have been collecting, which is held in disparate data sources. This has led to endemic sectoral anxiety and uncertainty, resulting in a paralysis of innovation. Local communities themselves might have much better ideas for effective services, but service providers often lack the funding and freedom to do things differently – and when they do, delivering and measuring impact is challenging. Data has the potential to inform communities about what an effective service might look like and to advocate for the funding of those services.

Accurate data is critical for successful service development, particularly for targeted, place-and population-based services designed to be more effective than blanket policy. Achieving this requires the collective will and freedom to share and blend multiple sources of data from multiple agencies, communities and not-for-profit organisations.

Data generated through the delivery of services is currently captured and held in isolation by individual organisations and government agencies – data that exists for the civic benefit of the community and its constituents rather than for any corporate or profit-motivated interest. As a unified conglomerate, this data represents much greater potential insight than its individual collections. Empowering individuals within communities with access to this insight poses the obvious challenge of the protection of privacy. The task of developing necessary robust and transparent data governance and protection processes is well under way. Innovation in this area is likely to come from the community itself, where place-based initiatives are central to local socio-economic prosperity.

Some philanthropists are already seeing the possibilities in this space, recognising that the current funding model perpetuates reliance on governments. One of Australia’s largest managers and distributors of philanthropic funds, Perpetual, is attempting to counter this by facilitating dialogue and providing capacity-building grants for not-for-profits. Caitriona Fay, Perpetual’s national manager of philanthropy and non-profit services, says not-for-profit executives see value in investing in digital and analytics infrastructure for service delivery. In July 2017, Perpetual will again host Stanford University’s Digital Civil Society Lab to foster sharing and debate among philanthropic and not-for-profit leaders. Perpetual’s clients have also made one of Australia’s first major grants ($1 million to The Smith Family) for building capability in collecting and using data, which includes investment in data science skills, development of new processes and the implementation of technology platforms to provide insights that will improve the delivery of services.

One example is now underway at Australia’s largest data innovation group, Data61, recently born out of the National ICT Australia research centre and CSIRO’s Digital Productivity and Services Flagship unit. The group connects and publicly releases disparate and wide-ranging government data sets for the benefit of industry. One current project of theirs is using data analytics to identify the most effective early intervention policies to address issues surrounding out-of-home care, including foster care, for more than forty-three thousand young Australians. The project aims to enhance service delivery, personalise services and improve outcomes for affected families. Access to the right data, as well as the skills and techniques for effective data interrogation, helps organisations design better solutions and interventions.

 

AUSTRALIA’S POLICY-MAKERS, IT would seem, are beginning to understand the importance of data and predictive analytics for administering an effective welfare system. In February 2015, the Australian Government released A New System for Better Employment and Social Outcomes, a report on the review of Australia’s welfare system, which describes the complexity of the policy and administration environment:

Changes to the system over time have led to unintended complexities, inconsistencies and incoherencies. They have created disincentives for some people to work.
The system is out of step with today’s labour market realities and community expectations.
It is failing to identify groups at risk of long-term income support dependence and needs to refocus on early intervention and supporting individuals through difficult transitions.
Without reform, the fiscal, economic and social sustainability of the system will be compromised.
A new social support system is needed to improve employment and social outcomes. The current system is complex and does not support everyone, who is able, to work and be self-reliant.

Local communities recognise this, particularly those in decline, distress or disadvantage – of which there are more than two hundred and ninety communities in Australia, according to research in the Place-based Impact Investment in Australia report from 2012. Shepparton, located in north-central Victoria and with a population of sixty-three thousand, is one of those communities.

The region has a long and proud agricultural history, but has weathered economic and social decline in recent years. The Greater Shepparton Lighthouse Project aims to do things differently to break the cycle of disadvantage. Research by the Lighthouse Project found that almost one in three young people (aged fifteen to twenty-four) in Shepparton were not engaged in work or study (the national average is around one in ten). Educators, community leaders, community workers and service providers describe these young people as ‘disengaged’. Understandable, when you consider communities like Shepparton now have five generations of unemployed people.

The reasons for youth disengagement in Greater Shepparton are complex, though there are several common risk factors including trauma in the home; breakdown of the family unit; physical, verbal and online bullying; anxiety issues (commonly linked to bullying); and truancy (also commonly linked to bullying). The Lighthouse Project takes a place-based, ‘collective impact’ community investment approach, modelled on similar strategies implemented to great effect in the US. Local community and not-for-profit organisations, government agencies and schools work together to understand where to focus efforts. The collective impact approach has been adopted by a number of communities around Australia, many of which are regional or remote communities. The funding for the design and co-ordination of the collaborative approach is typically made available through philanthropic or government grants. The missing piece for Shepparton is the ability to use data and analytics to understand the current situation better and predict likely outcomes, thereby helping to build a clearer case for early intervention.

Data and analytics are emerging as a new and important toolset for collective impact community organisations and Government agencies working together at the grassroots. Sharing data across organisations and agencies, and mining that data for insights, will provide, in the case of the Shepparton community, the evidence to enhance and complement understanding of when and why young people disengage.

The executive director of the Lighthouse Project, Lisa McKenzie, says, ‘This is definitely a piece that’s been missing from our work. We are yet to progress much beyond our initial state of Shepparton’s Children’s report in sourcing, holding, sharing and using data to drive change and this is a critical piece.’

In 2017, Lisa convened local decision-makers, community leaders and CEOs to explore how to use data and analytics to guide the change they wanted to see, examine specific implementations of data analytics for the social sector, and scope a pilot project to develop data tools for Greater Shepparton. This work has the potential to empower the Shepparton community with insights from the information they already collect, and help them advocate for the delivery of services they really need.

Opportunities for applying data analytics and data-driven technologies within the social sector are innumerable, and the solutions scalable. As many problems as there are to fix, there are data-driven approaches to help bring fresh perspective and deeper insights, advocate for and deliver better outcomes through targeted services, manage limited resources and design new ways forward. Why should these tools and technologies only be used to build staggering private wealth? Why should data about who we are be used only to recommend our next impulse buy? Data is everywhere, driving everything, and the social sector is the key to a future of balance and shared benefit.


From Griffith Review Edition 56: Millennials Strike Back © Copyright Griffith University & the author.

Griffith Review