Economics, Politics

Socialism is worse than capitalism—you want a welfare state

The rise to prominence of Jeremy Corbyn and Bernie Sanders has seen a return of ‘socialism’ to the vernacular. The political movements behind these two men have frequently sought to associate popular welfare policies, notably universal health care, with socialism. Both of them seem to be suffering from a degree of conceptual confusion between socialism and the modern welfare state.

This is ironic because it is precisely the socialist as opposed to welfarist elements of their respective platforms, like rent control in Corbyn’s case and trade restrictions in Sander’s case, that are unpopular. A second irony, one that is not lost on left-leaning economists, is that free markets, the antithesis of socialism, are necessary for providing the funding for a modern welfare state.

What is the distinction between socialism and the modern welfare state? One way to think about it is in terms of market intervention vs. post- and pre-market intervention. Market interventions are those that distort prices and inhibit their ability to communicate the opportunity cost of a good. A pertinent example is rent control.

The market clearing price ensures that homes are built until the amount renters are willing to pay equals the costs associated with building dwellings. If the market price is distorted so that it is cheaper than the market clearing rate, then there will be excess demand for housing but suppliers will be unwilling to meet it because they cannot cover their costs. The end result is a huge short fall in housing—not exactly a socially desirable outcome.

Gunnar Myrdal, a Nobel laureate in economics and an influential architect of the Swedish Labour Party’s welfare state, once quipped that: ‘in many cases rent control appears to be the most efficient technique presently known to destroy a city—except for bombing’.

A post- or pre- market intervention lets the market run to ensure the efficient allocation of scarce resources, but takes action at either end of that process to protect people from the ravages of the market. An archetypal example is the Danish flexicurity system.

The Danish labour market is relatively unregulated. In particular, you can hire and fire workers easily. This reduces the risk employers face when taking on labour, which encourages them to make investments, notably in emerging products and business expansion, which creates jobs.

This encourages an environment of Schumpetarian dynamism or ‘creative destruction’. As obsolete industries become unprofitable they can easily lay off workers and downsized gradually until they close completely. Meanwhile, new firms in new sectors, which are almost invariably more productive than old sectors, can pick up the newly released labour.

Schumpetarian dynamism entrenches a fast and efficient allocation of resources to their most productive use. This is the ‘flexi’ part of the system. The ‘security’ is where the welfare state comes in. The Danes recognise that an unregulated labour market gives people very little job security. To correct this social harm, the state pays almost replacement wages to recently laid-off workers while they find new jobs, and also provides job search and education services.

The flexicurity system combines the efficiency of free markets to increase the size of the economic pie and the effectiveness of the state transfer apparatus to achieve the humanitarian goals of left-wing morality. This is typical of modern welfare states.

Another outstanding example of the modern welfare state, in this case operating through a pre-market intervention, is the Australian system of income-contingent loans for higher education.

An income-contingent loan doesn’t need to be paid back until the debtor is earning above a certain level of income. Such loans are typically made to help people invest in something that will help them achieve precisely this level of income, such as education or migration. The higher education loan system is an ingenious welfare policy because it corrects a market failure and thereby allows that market to function more effectively, all while increasing equality of opportunity.

The market failure in this case is a ‘missing market’. Students are difficult to give loans to because they lack collateral and financial institutions are uncertain as to the likelihood that they will pay back loans. This is called ‘adverse selection’. As a result, financial markets  simply won’t supply enough loans to meet demand from students.

Yet at the aggregate level, most university students will go on to work, and at higher productivity thanks to their education. So as long as you get all of them into the system, the masses will make up for the few bad apples that default and you’ll have a better workforce to boot!

The state is the ideal instrument for this because it can track and manage all the repayments through the tax system.

Unlike pure state-financed free education, the income contingent loans system allows the price of tuition to reduce subscriptions to bad-value degrees. It is also transfers some costs from the state to education consumers, who derive much of the product’s benefits. Loans are income-contingent so that students are protected from sudden bankruptcy owing to a recession as they leave university, but interest is charged at the rate of inflation, so there is an incentive to repay the loan.

One last archetypal welfare state policy is the single-payer health care system. This involves a state monopsony, which uses market power to reduce the cost of health care. Individual private providers are then left to compete for the State’s custom, yielding further efficiency dividends. Savings are passed on to citizens.

What is neat about welfare state policies is the combination of the best of right and left wing thinking: economic literacy and human compassion. The right tends to fixate on market efficiency without paying due heed to the ravages of those markets. It is undeniable that markets can leave many, especially among the already marginalised, suddenly destitute, generally insecure and with limited prospects to achieve an improved situation for their children.

Yet it is undeniable that markets are extremely good at allocating inputs to where they will be most productive in producing output, thereby making us collectively richer. Socialist market interventions like minimum wages, industrial dispute regulations, price fixing, rent control, subsidies and of course, public ownership of the means of production, impede this efficiency and thereby reduce total output.  This is the origin of the quip: ‘socialism is an equal share of poverty’.

But Socialism does achieve equality. With large taxes and transfers and substantial regulation of market forces, people can be brought closer to absolute equality. With heavy state provision of infrastructure, health, transport and education, people are also given equality of opportunity.

The modern welfare state achieves both efficiency and equity. The market is left to run relatively freely, maximising income. Effort is rewarded and people are incentivised to work. The state then operates around the market to improve equity. It also looks to improve the functioning of markets in cases of market failure rather than simply replacing the market as in central planning.

It is sad that so much political commentary falls into unproductive comparisons of socialism and free market fundamentalism. There is a middle ground, exemplified by Scandinavia, Canada and Australia. It is the modern welfare state. The welfare state is not a choice between markets or governments, but a case by case approach to policy problems. Markets and the State each have their strengths in a policy context and need to be utilised judiciously.

The Obama administration borrowed heavily from the school of the welfare state, and Clinton seeks to build on the foundations it has provided. Sanders and the libertarians of the Republican Party, on the other hand, are advocating for only half of this efficacious system. It would be very beneficial if they could stop trying to trace their way through the maze of modern policy complexity by simply painting over it in broad ideological brushstrokes.


Mark Fabian is a doctoral candidate in economics at the Crawford School for Public Policy at the Australian National University.

Filed under: Economics, Politics


Mark Fabian is a doctoral candidate in economics at the Crawford School for Public Policy, The Australian National University. He blogs at


  1. > Socialist market interventions like minimum wages, industrial dispute regulations, price fixing, rent control, subsidies and of course, public ownership of the means of production, impede this efficiency and thereby reduce total output.

    Actually no. Labour markets are typically not in perfect competition are they? There is usually many sellers of labour and fewer buyers of labour. That means that the market is skewed in a negative fashion to the seller of labour (lower wages than what a perfectly competitive labour market would provide), with aggregate social costs (lower wages, lower productivity, greater unemployment) and with deadweight costs.

    > Not really; the gulf in command economies…

    Was actually substantially less measured against Gini coefficients, wasn’t it?

    • Mark Fabian says

      Dear Lev,
      Thanks for your comments. On your second point, perhaps you missed this paragraph?
      “But Socialism does achieve equality. With large taxes and transfers and substantial regulation of market forces, people can be brought closer to absolute equality. With heavy state provision of infrastructure, health, transport and education, people are also given equality of opportunity.”

      I did not suggest that markets result in equity. The whole point of the article is that you need to combine market efficiency with other measures to achieve equity. In the absence of market efficiency you have relatively less wealth to distribute in the first place e.g. West vs. East Germany.

      On your first point, let me first ask for some evidence. I would say that yes, they are usually in perfect competition. Then I would add that classic models of monopoly power, which are substantially empirically verified, suggest very few actors (~5) are required to achieve pure competition effects. There are some industries where this doesn’t come about, like aerospace engineering. But here labour is typically in short supply as well, so bargaining dynamics are actually quite even if not in favour of labour. People can also change industries, unless of course they live in a socialist country where the government allocates labour.

      The case that you might find most interesting is monopsonies in minimum wage industries, notably fast food in the United States. This is the site of a heated debated in labour economics since Card and Krueger published a paper arguing that, in the presence of such monopsonies, minimum wage legislation might actually increase both wages and employment by offsetting monopsony power. Their model seems good to me, and while people have contested their empirical findings (which were strictly for Pennsylvania if I remember correctly), I found them quite convincing. I stress though that this is a marginal case.

  2. P. brivio says

    I’m not an economistor clever like you but it seems to me say that your free Market provides better welfare than any other is dubious. What about those that can not play a full productive role in your Eutopia ? Who can not pay into it ? Disabled people are dying in the UK due to the harassment and Austerity cuts inflicted upon them by our government. The marginalised and disabled did not cause the Economic crash but they have to help to pay off the debt with less benefits and being labelled scroungers ! The wealthy have not suffered at all and in many cases got even richer due to their involvement in the sub prime crash but they receive no sanctions . No one wants a socialist state like you describe but a compassionate state where those less fortunate are given fair Involvement and use of the privileges of the capitalist system is not socialism it’s just morally right is it not ?

    • I’d give the essay another read, P brivo. Your comment about “your free market” shows you missed a lot. Google “welfare state” and then give it another read. You are more in agreement with the author than you realize.

  3. R. Schwartz says

    This article closely matches thoughts I’ve had for years. I’m really glad to see it.

    Many people who think themselves socialists actually are not. The core of socialism is democratically owned and run workplaces. So, by my view, you are not a socialist unless you think the government should pass laws that force some large portion of the economy to be owned and run by workers. Although, I do consider price controls, government ownership, and mandated unionization socialistic.

    If you think that the private ownership of capital should be permitted and prices should be determined by markets, then you are fundamentally pro-capitalism. Debates about how and how much to redistribute and regulate market failures are debates about the welfare state within a capitalist system, not socialism versus capitalism. I’m wondering if the author agrees with my formulation.

    This is the core insight I got from the article: “What is the distinction between socialism and the modern welfare state? One way to think about it is in terms of market intervention vs. post- and pre-market intervention.” I look forward to more from Fabian in the future.

    Here is a good discussion on the minimum wage research.

    Neumark is perhaps the leading US critic of the minimum wage. Since we know that people who don’t support the minimum wage have deep dark motivations—evil, hatred of poor people, profit, etc.— be aware that he is a Democrat and does support other measures to help the poor. (There is sarcasm in this last paragraph.)

    • Mark Fabian says

      Thanks for your comment. I do agree with your formulation. My view of the US minimum wage debate is here if you are interested:

      Long story short, the minimum wage is a welfare policy and needs to be delivered as part of a welfare package. If you just increase the minimum wage without appropriately changing the US welfare system you will simply see mass unemployment. Empirical papers cited in this area by proponents of the minimum wage are almost invariably from the era when the fallacy of the inflated denominator was being used i.e. measures were of the effect of changing the minimum wage on *average* employment, when of course you want to see the localised effects on the employment of minimum-wage labour. That localised effect can be huge. I know of at least one estimate that puts the price elasticity of minimum wage labour in Australia at 0.7, though we have a large pool of teenage labourers with poor wage rights to substitute to.

    • I think this is foolishness. The minimum wage is already so low that you can’t possibly live on it. What possible benefit would you hope to achieve by its elimination? Will a new era of prosperity ensue if someone at McDonald’s earns $4 an hour instead of $7.25? The only way minimum wage becomes a problem is if it is too high (we can’t force McDonald’s to pay the employees $50 an hour)

  4. “We need scarcely add that the contemplation in natural science of a wider domain than the actual leads to a far better understanding of the actual.” Sir Arthur Eddington – “The Nature of the Physical World”
    “The truth is rarely in the middle.” Joscha Bach — AI researcher
    Starting with those quotes because while this piece is well done and makes sense as fair as it goes; think that when one expands to a wider domain, capitalism and socialism have been both been blown up exponentially accelerating complexity. Not saying they’re evil systems. They made large and significant contributions to human well being. Think they’re time has passed.
    It ain’t rocket science: If your culture has deadly relationships with the sky and ocean, your cultural genome sucks.
    If your culture’s manner of reality interface is converting the sky into a lethal gas chamber, largely a gas chamber of commerce, that ain’t efficient. It’s time to revisit, rethink and redomain your cultural genome and the systems that genome builds, processes, etc.
    Redomaining is “ … the expressing of a given purpose in a different set of components …” Brian Arthur “The Nature of Technology.”
    Here are some examples: agriculture redomained food procurement from hunting and gathering; alphabet code redomained writing from pictograph code; steam technology redomained power generation from water wheel technology; electricity redomained power generation from steam power; democracy redomained government from monarchy; eukaryote cells redomained biotech from prokaryote cells.
    I study code in a physics / evolution context, including monetary code.
    Have a new thought-structure regarding how human-as-cells in cultural organism interface with reality.
    Too much to delineate here. Here’s a link to an abstract of sorts:,_Complexity_%26_Code.html
    (It’s an an abstract for Complexity, Culture & Code, a longer piece that I queried Quintette about but have yet to hear back.)
    Naturally, being an abstract, it’s incomplete as well, but it does provide at least part of the philosophic foundation for redomaining our cultural genome. Myriad implications . . .

    • Mark Fabian says

      Thanks for your comment. There is actually quite a lot of work being done on social philosophy in a post capitalist/socialist liberal/fascist space, notably in the institutional literature and in public policy discourses. I tend to dislike the work coming from scientists in this area because they commit pretty basic errors owing to their general ignorance of the history of ideas in political, social and economic philosophy. Perhaps this is a by-product of being used to laboratory conditions, when social scientists are more awake to the messiness of social science. For example, one notion that I come across *all the time* is that we should conceive of a perfect society and then figure out how to get there, rather than simply focusing on fixing problems we can identify in the present and thereby arriving at a better society iteratively. This is one of the most prominent themes in social philosophy from the 20th century, notably in Popper’s work, which is unsurprising given that the 20th century was punctuated by our last attempts to jump to utopia: communism and fascism. Any democracy theorists can confirm that we are better at agreeing on what is bad than on what is good, and it should not be hard to convince someone with liberal inclinations (e.g. pluralism) that negative utilitarianism is a much safer and more ethically defensible paradigm for a state to use that a positive moral code. Scientists also tend to ignore the basic tenet of Burkian (i.e. sensible) conservatism, namely that the system we have now has evolved slowly to its existing level of extreme complexity, and any attempt to change it is likely to have unforeseen and potentially dramatic consequences, so we should tread carefully.

      With regards to paradigm shift to address environmental concerns. I am unconvinced that a system change is required. Many of our existing tools can fix the underlying problems, like a lack of regulation and pricing for the global commons, pollution etc, and we are rapidly coming up with new tools for others (e.g. global treaties, the G20 etc). I certainly think that over this century we will start to see a coalescence of ideas around new modes of social organisation (automation and the digital age seem likely to make that inevitable), but I think it is still a way off.

    • Mark Fabian says

      Just read your abstract. How do you see ‘software code’ replacing ‘monetary code’, by which I take it you mean price signals, as a means by which to ‘respond with sufficient reach, speed, accuracy and power to generate functional relationships in and across eco bio and cultural and tech networks, and across time’? Are you basically just arguing that big data analytics can tap tacit knowledge networks better than price signals can?

  5. Hey Mark,
    Thanks for your comment.
    I’m not talking about utopia, or one size fits all, some monolithic perfect society construct. Impossible. To drop me in that slot is inaccurate, but this medium is very limited, so no worries.

    Redomaining is a process. Just because you move from monarchy to democracy (or prokaryote to eukaryote) doesn’t mean all your problems go away. Now you have new problems. The point is that the information processing has moved from a King who needs a nap to a vastly larger network of information processors with myriad interests and abilities who can better navigate the new complexity of the industrial revolution. The information processing reach has expanded tremendously.

    Think that is the essence of why capitalism kicked communism’s butt. 50 committee members doing a 5-year plan for shoe production, trying to figure demand, colors, etc. Ouchness.
    As you know, the information processing mechanism of the market’s thousands of producers and millions of buyers processing shoe-relationship information in shorter intervals is faster, more accurate and more powerful than 50 bored guys trapped in a room for 2 months. As David Deutsch says, systems needs rapid error detection, error detection mechanisms. The market did that much better a planning committee. It wasn’t close.
    Brian Arthur uses a surfing analogy regarding navigating complexity. It’s like surfing in that you’re not going to control the wave; you ride, the wave changes, you react appropriately; repeat.

    Information processing, that’s the essence of survival, and what I’m talking about with the move from humans using monetary code to thousands of quantum computers, add IBM’s Watsons’ everywhere using software code, ubiquitous networks sensors collecting big data from rivers, forests, an internet of things, etc., analogous to an immune system constantly monitoring the body’s various networks; or building a technological neocortex for culture.
    What I’m talking about is applying the following rule of thumb from the biological network to the cultural network.
    “The rule of thumb is that the complexity of the organism has to match the complexity of the environment at all scales in order to increase the likelihood of survival.” Yaneer Bar-Yam “Making Things Work”

    I disagree with you significantly about the state of our environs. The methane is bubbling; there’s geologic precedent for Abrupt Climate Change; rapid devastating increases in temps. Paul Beckwith, a reputable climate scientist reports that we’ve entered abrupt climate change. He’s still trying to bang on the drums of cultural evolution … and as other knowledgeables state: collapse appears increasingly likely. You may have seen the science study that says we’re crossed 4 of 9 planetary boundaries.
    I interviewed Dr. Frank Vertosick; he thinks the idea that humans can be planetary stewards is just more pathetic human hubris; seems more and more correct. (damn . . . where’d I put those painkillers.)

    From my perspective (and others) think it’s Triage Time, or already too late.
    “The point is that underfire, the best strategy is to try everything. We can’t wait for Plan A to fail before we initiate Plan B; this approach takes too much time …” Rebecca Costa — “The Watchman’s Rattle.”
    Replacing monetary code? Absolutely preposterous, ridiculous politically, etc. “Shut! Up! Dude!” I get that.
    But we’re talking survival; and argue this is about the science of information processing so our cultural genome matches the complexity of our environs. Any political changes flow from that.
    We could try software over monetary code in a small area, And as Dr. Costa wrote, try 50, 500? other ideas in other areas, other countries, like an immune system strategy. It doesn’t know what’s going work, generates variation, processes feedback, reacts appropriately; repeat.
    We don’t know what’ll work. I sure as hell don’t, but my strong sense is that incremental edits of policy and pricing are hopelessly inadequate processing mechanisms. Exponential is a wild, unwieldily beast that I don’t think we’re biologically set up to handle, or culturally right now. Computers can, or could.
    Okay, too long.
    On a human level, I appreciate your evident humanity Mark. Appreciate that you have knowledge I don’t have, your classy restraint (a strength; and a weakness in certain situations).
    Know I’m full-of-wrong, by def, and try to retain a willingness to be wrong.
    Hey, I hope you’re right: we have time. My processing doesn’t lead me to that conclusion.
    Regardless, Best to Thee . . .

    • Hey Mark, we posted almost simultaneously. Hope the above answers your question . . . hey, me cognitive bandwidth be fading.
      Also, mistake. My statement: cultural genome matches the complexity of our environs . . . should have been: cultural genome matches the expanding reach of culture and the increasing complexity of the new relationships that expansion generates.

    • Thanks for this long comment Bryan, and for your generous compliment in closing. Same to you. I like your ideas and I certainly think the collapse view is very plausible, I’m just not *that* worried. I certainly think data analytics have immense potential for the kinds of things you are talking about, but replacing monetary code as you call it is a bit premature. Replacing monetary code for these specific problems seems mostly fine, but I think you’re still going to need price signals for anything of economic substance somewhere in policy design. In any case, please send me a link to your paper when you publish it. My ANU email address is publicly available. Best wishes.

  6. Great article, but I think equality of outcome is a red herring and is one of the aspects of the Left that has to be ditched.

    All you want is an absolute floor or safety net, and that no-fuss guaranteed, a decent (not uncomfortable, but not comfortable either) flat basic income for individuals qua individuals that one ceases to get above a certain threshold of income where it becomes comparatively negligible, at which point one starts paying taxes, starting at 5% and incremently increasing to about 15% or so.

    A welfare system should also include a plethora of re-training options in all sorts of skill areas. This is in fact the most important aspect of any proposed welfare system and it needs to be designed very carefully to avoid “cowboys”, free riders, etc.

    This does have a certain degree of inefficiency from a market point of view (it removes the fear of absolute poverty and death as an incentive), but obviously the moral argument plus the other positives on the economic side that you mention (Schumpeter) outweigh that.

    Medical care needs to be a bill that you pay and own, that’s paid for by a loan from the state that the state (as it were) doesn’t care all that much if you pay back – i.e. pay it back as and when you can, in tiny amounts, and maybe if it was a big bill you’ll never be able to pay it all back, but who cares, so long as you make some effort – and you are always going to be gently reminded to make the effort. Again, beyond a certain income, this falls away.

    Those things, plus some basic infrastructure and defence, and some basic core legal and policing functions (not all, most stuff in this area can be handled by the market too) is all the state needs to do, the rest – subsidising industries, control of money, most infrastructure, most law and policing,arts, yadda yadda, it all has to go.

  7. Pingback: Socialism, globalization, populism | Rturpin's Blog

Comments are closed.