[This is a post in my ongoing series on problems I see with libertarianism as is. There are many parts that I would defend. However, my take is that just too much is wrong, and that libertarianism has to be fixed in many ways or replaced with something else. In addition, many problems are neither new nor marginal. You can find an overview of all my articles here that I will keep updated: “Synopsis: A Critique of Real-Existing Libertarianism.”]
Let me start on a personal note. This is somewhat of a lenghty digression before I address the main point, but I hope it is interesting in and of itself.
In mid-2014, I did a short presentation on free migration for a conference of the European Students for Liberty. Afterwards someone suggested that I should also talk about the topic at one “Hayek Club” close to where I live. It took until 2015, though, before the contact was made. I proposed several special subjects, one of which was “Immigation and the Welfare State.” That seemed to strike a chord. I then tried to explain the thrust of what I wanted to speak about to the person from the “Hayek Club.”
With low immigration as under current conditions (about 0.3% of the resident population annually for the US and the EU), any effect would have to be minimal. Even with considerably higher immigration, as for example in the 19th century (about 0.9% annually for the US), it should not be materially different. Apart from that, there are arguments on both sides that are apriori plausible: Immigration might lead to a higher share of the welfare state or a lower one. Different effects could cancel out, and so my hunch is that the net outcome might be close to zero with an unclear sign. This is an empirical question because the question cannot be settled apriori.
— — —
What I noticed was that the guy from this “Hayek Club” seemed to be disappointed with my argument. He had apparently expected something else. I checked his Facebook page and noticed a certain slant there. My suspicion was that he was ideologically close to the rightwing populist Alternative for Germany. And so a safe bet was that he thought I would prove that immigration leads to catastrophic results. Note that that was before the large influx of refugees started in late 2015.
What I noticed was not a one-off at the time. A little later a rift within the “Hayek-Gesellschaft” (Hayek Society) erupted that also this “Hayek Club” is associated with. Many people noticed what they interpreted as entryism from the far right and a takeover. They left the “Hayek-Gesellschaft” in protest and denounced the drift to the Right. If you can read German, here is one article of many about the development. Actually, there were similar symptoms also elsewhere at the time. When I look at the webpage of this one “Hayek Club” now and the people they like to invite, I would say that the party line is that of the Alternative for Germany. It is just presented as “classical liberalism” and “libertarianism,” which makes me cringe.
— — —
In the discussion with this person from the “Hayek Club,” certain arguments came from his side. The underlying belief was that there was an inevitable and fast expansion of the welfare state underway that would be sped up dramatically by immigration. I tried to argue against both claims. However, it seemed impossible to get through. One argument that I made was the one I will explain below, which, in my view, contradicts an argument about an automatic expansion of the welfare state so patently that at least some pause should have been warranted. Still the claim was treated as axiomatic by this guy from the “Hayek Club,” and not in need of an empirical check. I felt like talking to someone in a cult.
As I insisted on my arguments, the initial enthusiasm about an invitation faded away. There was no initiative to make it happen. I kept on asking, and then it was supposed to take place almost in secrecy on short notice just a few days before. I asked whether the “Hayek Club” was really interested, which was obviously not the case, and then I canceled the presentation because it seemed pointless and a complete waste of time.
— — —
Those who left the “Hayek-Gesellschaft” claimed that the move to the Right was a betrayal of Hayek’s ideas. In a way, I think this correct. But all in all, I don’t think it is completely true. The claims about an automatic expansion of the welfare state, and by extension a transition to a full-blown planned economy and an authoritarian regime, are vintage Hayek. He was not the first to make such a point. Also for his teacher Ludwig von Mises the assertion is axiomatic. Any “interventionism” is just a slower transition to a quasi-Communist system.
Hayek develops the argument in his “The Road to Serfdom.” He wrote the book between 1940 and 1943, and it was published in early 1944. I do not understand why it is viewed as a clairvoyant critique of Fascism and National Socialism. The former had been around for two, the latter for one decade. This is as impressive as if I wrote a book now on how Facebook will play a huge role, or even the Internet. “The Road to Serfdom” was actually a late entry to the discussion just before it was too late for predictions in 1945.
Hayek, just like Mises in his “Omnipotent Government,” also published in 1944, basically interprets Fascism and National Socialism as examples of economic planning. I have written about why this is at best very incomplete in my post “Were the Nazis “Socialists?”.” What both Hayek and Mises mostly ignore, though, is that Fascism and even more so National Socialism had an ideology and worldview that to a large extent had nothing to do with Socialism or economic planning. The elephant in the room: race theories, eugenics, the quest for “Lebensraum” in the East and killing off millions of people, hardly played a role in Hayek’s and Mises’ books.
With such a distorted view, I find it little surprising that Hayek’s and Mises’ predictions were also spectacularly off. Mises thinks that Germany would probably see a Communist uprising. To counter a Germany that would remain problematic, only the concept of “Intermarium,” a federation between Germany and the Soviet Union from the North to the South through Eastern Europe, could work. I am willing to grant here that it was hard to foresee the future in 1944. But then this is not an indication of deep insight either.
Hayek had his own theory. While faced with still virulent and powerful National Socialism, his main concern was that a planned economy in Great Britain during the war would set the country on an automatic “road to serfdom.” He then supplies many arguments why interventions in the economy may propel themselves forward, mostly through ideological and political channels. It is somewhat inconsistent to argue against what you present as almost deterministic because this implies that the outcome is open and can be swayed by rational argument. But then Hayek can hold both views at the same time.
— — —
Although, libertarians can read “The Road to Serfdom” as visionary, Hayek’s forecasts actually also turned out to be false. Indeed, Labour was determined for the next decades to expand the role of the government in the economy. There were nationalizations, the National Health Service came in 1948, and there were further welfare programs. However, the predicted dynamics did not materialize.
Basically, Labour drew their inspiration from the Fabian Socialists, not the Marxists. The Fabians had an ambiguous relationship with authoritarianism, but still developed from a liberal starting-point. That probably made a crucial difference: Great Britain remained a liberal democracy. The Tories took over again already in 1951 and for the rest of the decade. Labour was only intermittently in power afterwards and actually only for about twelve years all in all from 1945 until 1979 when Margaret Thatcher became Prime Minister.
While Labour could implement part of their program, and the Tories were not completely averse to welfare programs either, the net effect never came even close to a fully planned economy although the welfare state expanded to some extent. Also predictions about an inevitable move to an authoritarian order did not come to fruition. Or in other words: Hayek was wrong.
Hayek could perhaps have fixed this contradiction with a story that Great Britain only got around the great danger because of fortunate circumstances. Maybe it was the valiant fight of Margaret Thatcher. However, she became Prime Minister only in 1979, after more than three decades. Would that not have been too late if Hayek’s theory were correct?
And then the predictions also failed for other liberal democracies. The Social Democrats in Sweden were much longer in power. It did not happen there either. Actually, it almost never happened. At best, you could point to Venezuela where Hayek’s prediction has roughly played out. But then you do not need some “why the worst get on top” explanation à la Hayek in this case because Hugo Chávez was in charge from the get-go.
Even if you grant this example, Hayek was overwhelmingly wrong. And then there are far more examples of liberal democracies turning sour for completely different reasons and with other ideologies mostly from the Right in the background. Maybe that should have been Hayek’s main focus. But if you have an anti-Socialist hammer, everything looks like a Socialist nail.
— — —
Hayek fans are probably unfazed by this observation, and that was also the case when I pointed it out to the guy from this “Hayek Club.” One way to mend the cognitive dissonance is to note that the amount of money going to the government, including the welfare state, has grown over time. But absolute numbers are not the point here. With a growing economy the amount of money can increase, even while the percentage of the economy goes down.
What is relevant is the relative share. Now, that has also gone up over time in practically every developed country. In 2012, the share of GDP under control of the government (“Staatsquote” in German) for EU countries ranged from somewhat more than 40% for Ireland, Spain, and Luxembourg to 58% for Denmark (data here), with an eyeballed mean in the upper 40s. For the US it stood at 37.3% and for Switzerland at 34.7%.
However, lean back and let that sink in: The difference was not as stark as many Americans, especially libertarians, assume. It is common to contrast the US with the EU as “Capitalism versus Socialism.” Surely the share of government was lower in the US, but not by that much. If the claim made sense, you should expect something like 10% versus 90%, not 37.3% for the US versus 44.9% for Germany. This is a difference in degree, not in kind.
— — —
One way to salvage Hayek’s prediction is to point out that was wrong about the speed, but got the tendency right. Still, if roughly the share of the government stands at about half the total economy, this is not a fully state-run economy, only a mixed one. If you take the long-term increase, and extrapolate as if that were warranted, it would probably take a few centuries to reach a full-blown planned economy.
But then this is not warranted. One clue here is that the share of the government in the economy falls in a rather narrow range for different countries. If they had pursued different random paths with a drift upwards, the spread should probably be much larger. You would then perhaps also find some countries with shares far above 50%. However, as far as I can tell, none of them ever got past 60%, few made it past 50% by a large margin.
The other curious point is that the claim of an automatic expansion of government activity, including the welfare state, is at odds with the data. Here is the development for Germany:
I will skip the inconvenient fact that the government share was way higher during World War II, as far as I know well above 70%. In 1960, it stood at 32.9% (data here). It then went up like this: 37.1% in 1965, 38.5% in 1970, and 48.8% in 1975. So Hayek was right after all?
Well, no, because here is how the data series continues: 46.9% in 1980, 45.2% in 1985, 43.6% in 1990 (note that reunification occurred around that time), 48.2% in 1995, 47.2% (or 54.9% including debt resulting from the former East German economy, but only for one year), 47.6% in 2000, 46.9% in 2005, 47.7% in 2010.
In other words, what you have is a major ramp-up from 1960 to 1975, and from then on the share of the government remains essentially flat for 35 years. This is not the behavior that you would expect with an automatic drift upwards. It looks like something that grows to an equilibrium and then fluctuates around it. As far as I know, the pattern was similar also for other countries: A rather fast ramp-up and then flat-lining for decades.
How much of this was driven by the welfare state in a narrow sense (“Sozialquote” in the data series)? It went up from a share of 11.2% in 1960 to 17.7% in 1975, but from then on the increase was rather marginal, only to 19.6% in 2010. The share of the government in the economy was mostly not driven by the welfare state in a narrow sense although that played some role.
— — —
How to explain this?
Hayek’s explanation appears to be wrong. I would say that is so because he only focused on mechanisms that can increase the share of the government in the economy. He was looking for confirmation. But then there might also be mechanisms that go in the other direction and work against an expansion and check it from some point on. If that is so, the eventual level might be more like a constant, and a constant well below 100% for developed countries, a far cry from a fully planned economy.
My guess is that the missing link here is a functioning liberal democracy. Surely, welfare programs and other government activities are popular with large constituencies and that can lead to an expansion. But then there are also a constituencies that have a different take. They don’t want to pay ever more taxes. The larger the share of the government in the economy, the larger also the latter constituencies. If you have a functioning liberal democracy, where parties have to reckon with the anti-expansive constituencies, this should put a lid on how far it can go.
How to explain the pattern over time?
A plausible idea in my view is this: The German economy grew at a pace of 8.2% annually in the 1950s. That went down to still respectable 4.4% annually. And after that the rate of growth slowly leveled off with 2.9% in the 1970s, 2.6% in the 1980s, and somewhere between 1.5% and 2% ever since (data here, p.2).
Now, suppose that the government initially taxes the economy at some rate R. When the economy grows and the government taxes increases at some rate S. Over time, a larger and larger part of the economy will be the result of increases and is taxed at the rate S. No matter what the initial rate R was, the total rate will slowly converge to S, and that is the share of the government (I treat everything as a tax although in reality it might be under different headings.) Now, if the government takes 50% out of the increase, its total share will converge to 50% over the long-run. If it takes 40% or 60%, the asymptotic share would be 40% or 60%, respectively.
When the economy grows fast, there are what would be pithily called “Verteilungsspielräume” in German (roughly: leeway for redistribution). The government might take a large bite out of the increase and grow its share in this way. Even with a high tax rate, there would still be also rather fast growth after taxes, not as large as with a lower rate, of course, but still large.
For example, if taxation on the increase is 50%, and growth is 4%, then after taxes, it is still an increase of 2%, or a doubling about every 35 years, not too bad. If you assume “Friedman’s Law,” ie. that the government produces goods and services at about twice the price on a market, what the government provides for the taxes still has some value. So actually, the perception with 50% taxation on the increase might feel more like 3%, not 2%, or a doubling every 18 years. Worse than with lower taxation, but still not to be scoffed at.
During times of fast growth, and maybe also with a certain delay somewhat later because the political process may take some time, the government has leeway can expand its share. Surely, taxpayers would find it better without an expansion of the government share. But the pain is limited because you still experience an increase directly after taxes plus an increase through government good and services. Under these circumstances, an expansion of government activity is easy for politicians, and maybe to be expected.
However, with lower economic growth that changes. If you have only 2% growth, and you would still like to have your 2% increase after taxes, then you should be against any taxation on the increase. That would keep the size of the government absolutely fixed, while the rest of the economy grows past it and share of the government shrinks asymptotically to 0%.
Of course, that need not happen. There are different constituencies here that have different agendas and who have to work it out. However, the leeway for a further expansion of government activity shrinks with slower growth, or may even become negative. In other words: Mechanisms that work against an expansion start to kick in. At some point there will be an equilibrium that may depend on various other factors, but should still be well below 100%. As it seems, a typical range is 45% to 50%.
There can be additional developments: Debt financing can provide more means to the government, but that only postpones taxation to a later point in time. Or there could be on-offs like the now obsolete “peace dividend” after 1989 because of less spending on defense. But those mechanisms cannot work over the long-run. There may be some fluctuations hence, but only around an equilibrium level. That could then explain the long-run stagnation in the date for the share of the government.
I don’t know whether my explanation is correct, but it is broadly in keeping with the data, and Hayek’s underlying hypothesis is not. So it at least seems like a better candidate.
— — —
How could Hayek get this so wrong?
If you have a worldview that a welfare state will inevitably lead to a fully planned economy with the probable authoritarian implications for the political system, nothing can faze you. That’s perhaps an article of faith. I am for freedom of religion, but the claim is still contrary to the evidence.
I think there is an explanation why Hayek took an inevitable road to a fully planned economy for granted. It is funny that he agreed on this point with his opponents. Only that he saw this as ominous while they viewed it as promising. I have to go back in time to explain the point.
The initial Socialist program of Lassalle and even more so of the Marxists was that a fully planned and nationalized economy had to be the goal. When Marxism became the obligatory ideology of the Social Democrats from the 1870s on that was their stated program. Marx and Engels thought it impossible that there would be a peaceful and gradual transition to Socialism. It could only be with a revolution that would sweep the old system away.
But it did not happen. Marx and Engels kept predicting that the revolution was around the corner for half a century. August Bebel, the leader of the Social Democrats, was also always optimistic that the “große Kladderadatsch” (the great helterskelter) come about. But by the 1890s, many Social Democrats had lost their faith in his statements and started to ridicule him. As long as Friedrich Engels lived, though, the revolutionary way forward was a given. However, even Engels relented in his final years and conceded that it might also be a gradual transition. My guess is that that was not because of some insight after half a century, but because of the overwhelming evidence to the contrary.
Right after he had died in 1895, Eduard Bernstein who had closely worked with Engels challenged this view. He had probably had doubts for a long time about the official Marxist line. Bernstein lived in exile in Great Britain. And according to Engels’ predictions from the 1840s on, the revolution was supposed to break out in the most developed country first, and that was Great Britain. But if a revolution in Germany seemed improbable, it was unthinkable in the UK.
In addition, Eduard Bernstein had probably studied the Fabian Socialists who were highly critical of the Marxist dogma. One of their ideas was what was later called “Guild Socialism,” an economic system that was supposed to grow out of the co-operatives movement. This would lead to a “socialization of the economy,” not in the sense of state-planning, but as some kind of self-organization of society. The concept was vague, but it owed much to an underlying liberal Zeitgeist in the UK that was averse to an intrusion by the government into everyday life.
Doubts about the Marxist ideology had been latent also in the Social Democratic party for long. In 1890, the classical liberal Eugen Richter of the Freisinnige Partei published his “Die Irrlehren der Sozialdemokratie” (The Fallacious Teachings of Social Democracy), which sold several ten thousands of copies. In his book, he dissected the Marxist program of the Social Democrats. Richter was fine with the Democratic part, but his analysis of the Socialist part showed that a planned economy would be a disaster and a threat to liberty. The Social Democrats were outraged, but also short on arguments.
In 1891, Eugen Richter then published his “Sozialdemokratische Zukunftsbilder” (Social Democratic Pictures of the Future, but translated to English as “Pictures of the Socialist Future”, also available online). The book sold more than a quarter million copies in Germany, and it painted an eerily prescient picture of how Communism would work in the 20th century.
The Social Democrats dismissed Richter as a Capitalist lackey who wanted to preserve the status quo, which was silly because Richter was in the opposition all his life. Some Social Democrats, though, began to think about throwing Marxism out the window already in the early 1890s. Still the consensus held. Richter then cornered Bebel in debates in the Reichstag [national parliament] in the mid 1890s where it became transparent that Bebel had no idea how Socialism would work.
From 1896 on, Eduard Bernstein now started to publish as series of articles in a Social Democratic magazine, and in 1899 a book on the same topic: “Die Voraussetzungen des Sozialismus und die Aufgaben der Sozialdemokratie” (The Preconditions of Socialism and the Tasks of Social Democrady, translated as: “The Preconditions of Socialism”). Ostensibly, it was only a reevaluation of some points in Marxism, but effectively it amounted to a refutation of its central tenets, eg. the revolutionary transition to Socialism.
Bernstein’s vision was close to that of the Fabians. He was enthusiastic about co-operatives, which had long been despised by the Social Democrats as a mere palliative and distraction from the class struggle. Even worse: He was full of praise for Liberalism and viewed his take as a continuation, not an anti-thesis. Bernstein’s contributions led to what became known as the “Revisionismusstreit” (debate over Revisionism) within the Social Democratic party in the late 1890s.
What is notable here is that Eugen Richter supported Bernstein’s side in his “Freisinnige Zeitung.” In his earlier books, he had called on the Social Democrats to dump their Marxist program, and now it looked like that could happen. Bernstein never mentioned Richter, but I think he and his critique were in the background here as well. But then the Social Democrats officially hated Richter, so Bernstein could never have admitted an influence.
Eduard Bernstein’s vision was not classical liberal, but far more liberal in many ways than the Marxist party line. Basically, he did no longer challenge a market economy. It should only be “socialized” by a peaceful and gradual process of more and more social versus private production. “Social production” here was understood as by co-operatives.
The “Revisionismusstreit” ended with a de facto victory of the orthodox Marxists for the time being. The split in the party was somewhat papered over by August Bebel. However, in the longer run, Bernstein was successful. When the majority Social Democrats backed a liberal democracy after World War I and rejected the authoritarian model of the Bolsheviks, they fell back on Bernstein’s Revisionism.
That meant that they adopted his vision that in a liberal democracy, “socialized” production would simply outgrow private production. With additional welfare programs and some direct involvement by the government, this would lead to Socialism, ie. social production, not necessarily a planned economy, together with a liberal democracy at the same time. The Social Democrats now firmly rejected a violent revolution and a dictatorship of the proletariat as in the Soviet Union, while they had been ambiguous about these points of the Marxist program in Imperial Germany.
This Revisionist program was the vision of those on the Left, the German Social Democrats in Weimar Germany and the Fabians who inspired the British Labour Party. They took it for granted that the expansion of co-operatives and other reforms would result in a full “socialization” of the economy. Nationalizations and welfare programs were also on the agenda, but not a revolutionary change, but as a gradual expansion of the welfare state. The expectation was that this would eventually go to 100% of the economy.
And that is the context of Hayek’s criticism. The funny part here is that he took the assumption of his opponents at face value that there was an inherent tendency in a liberal democracy towards Socialism.
If my arguments above are correct, though, then both Hayek and his opponents on the Left were wrong about this. The enthusiasm about co-operatives faded away over time as they became a larger part of the economy, but remained only a part, and more so later when their importance declined. This led to a certain appetite for more direct government activity, but the mojo for a fully planned economy was by then used up with the failure of Communism in sight. All in all, there was an upside for a larger government share here. But only to a point, not to 100%.
— — —
The expectiation of both Hayek and his opponents on the Left was that the increase of the government share could be extrapolated to 100% in the future. It looked good until the 1970s. But then growth rates receded, and with them also the leeway that had existed before. So the share of the government leveled off at a much higher level than in earlier times, but well below 100%.
This should have enticed Hayek to rethink his claims. But as far as I can see, it didn’t happen. Libertarians have been stubborn about this for decades and in the face of overwhelming evidence. But then you can always fix a problem with “epicycles,” adhoc explanations as you need them. The favorite such “epicycle” is that superhumans like Margaret Thatcher halted the inevitable march to a planned economy, or Ronald Reagan did it, although both only presided over a stagnation, not a serious rollback of the share of the government in the economy or the welfare state more specifically.
The hilarious part here is that a very similar cognitive problem also set in for Hayek’s opponents on the Left. Their assumption was essentially the same, only with a different take that it was good: Capitalism would evolve into Socialism with some help from politicians on the Left. But it did not happen. At some point, it just stopped. The “epicyles” to fix this cognitive dissonance were parallel only with the opposite moral judgment: Villains like Margaret Thatcher and Ronald Reagan had managed to sabotage the inevitable march to Socialism. Rather mythical “monetarists” were at first the culprits who had engineered all this, but the word was perhaps too awkward. So the common explanation is by now that a neoliberal conspiracy did it.
In both cases, I would views this as an attempt to explain a refutation by the evidence away. It went against a central tenet of the respective worldviews that would have to be rethought with a concession. But people rarely do this in such a situation. If the facts do not fit the theory, change the facts. Hayekians can diagnose a massive increase of the welfare state, against all evidence of a stagnation, and those on the Left manage to find a massive decrease.
— — —
To sum up: Hayek’s prediction that liberal democracies would inevitably degenerate into authoritarian planned economies has not come true. The reason for this is probably that Hayek only looked for mechanisms that confirmed his worldview and ignored mechanisms that went in the other direction. As it seems, the share of the government levels out somewhere in a range from 45% to 50% of the economy, with some variation around it. The moderating influence here probably works through a liberal democracy, which seems like a bulwark against a planned economy.
Hayek possibly inherited his view from his opponents on the Left who shared his assumption that there was such an inevitable tendency to an expansion of government activity, although they held it for completely different reasons and with the opposite judgment attached. Both views run against overwhelming evidence, but that does not have to faze anybody. Hayekians reinterpret actual stagnation over decades only as a fortunate development because of heroic people like Thatcher and Reagan. Their opponents have their own fix and paint them as sinister figures that executed the program of monetarists or neoliberals.