You should be using margin when young, proof inside (serious advice/DD)

Not a joke, not meming. The advice to avoid margin around here is extremely foolish, and I intend to prove it to you in this post.

I think the most common retort I hear around wsb and online investment spaces is that using margin is obviously reckless, while “just investing what you have” is prudent, because you can potentially get wiped out and end up owing money. This framing is backwards and actively harmful because it is overweighting short term horizon risk.

But we're investors, and the goal should not be to minimize the chance of a scary year, it should be to minimize your lifetime tail risk. I.e, the risk of ending up with a disastrously bad outcome after an entire investing career then refusing to use any leverage when young can actually be the riskier choice.

The reason is simple: a young investor’s problem is not mainly “how do I avoid volatility this year?” It is “how do I spread market exposure efficiently across my entire lifetime?” The classic Ayres–Nalebuff lifecycle argument is exactly this: young investors are often underexposed to equity risk relative to their lifetime balance sheet, and moderate leverage early in life can improve diversification across time, not just across assets. In their historical and simulation-based work, they argue that leveraging equities when young and then reducing risk later can dominate both standard target-date style glide paths and a constant 100% stock allocation in retirement-outcome terms.

The key idea is that the relevant balance sheet is not just your brokerage account. It is:

Wttotal=Ft+HtW_t^{\text{total}} = F_t + H_tWttotal​=Ft​+Ht​

where FtF_tFt​ is financial capital and HtH_tHt​ is human capital, the present value of your future labor income. Lifecycle portfolio theory has emphasized for decades that young people typically possess little financial wealth but a large stock of human capital, and that this fact should matter for optimal risk-taking. In standard formulations, human capital often behaves at least partly like a bond-like asset, because it is a future stream of labor income rather than a traded risky portfolio. That is one of the main reasons many lifecycle models imply that the young can rationally take more financial risk than the old.

That observation leads to the first core equation in the argument. Suppose your desired exposure to equities is a fraction sss of your total economic wealth, not merely your currently invested savings. Then the dollar equity exposure you “should” want is:

Et∗=s(Ft+Ht)E_t^* = s(F_t + H_t)Et∗​=s(Ft​+Ht​)

But your actual unlevered portfolio can only put at most your current financial wealth into stocks:

Et≤FtE_t \leq F_tEt​≤Ft​

So when you are young and Ht≫FtH_t \gg F_tHt​≫Ft​, you are often massively underexposed relative to your lifetime balance sheet. The gap is:

Exposure Gapt=s(Ft+Ht)−Ft\text{Exposure Gap}_t = s(F_t + H_t) - F_tExposure Gapt​=s(Ft​+Ht​)−Ft​

If FtF_tFt​ is small and HtH_tHt​ is large, that gap can be enormous. In other words, a 25-year-old who is “100% in stocks” is often nowhere near “fully exposed” on a lifetime basis. They are only fully exposed relative to their tiny current account. Relative to total lifetime wealth, they may be severely underinvested in risky assets. That is the intuition behind leveraging early: not to become recklessly aggressive, but to reach a more even lifetime risk allocation.

Now I hear you -- you're asking me, but ol pencil of quantums -- Why does that matter for tail risk? Because long-run investment outcomes are path dependent. If all your real exposure to equity risk is concentrated late in life, then your retirement outcome becomes highly dependent on what the market happens to do in a much narrower window. You may be “unlevered,” but you are accepting a different form of concentration risk: time concentration.

A clean way to see this is to write terminal wealth as the compounded value of all contributions:

WT=∑t=1Tct∏u=t+1T(1+ru)W_T = \sum_{t=1}^{T} c_t \prod_{u=t+1}^{T} (1+r_u)WT​=t=1∑T​ct​u=t+1∏T​(1+ru​)

where ctc_tct​ is the amount invested at time t and rur_uru​ is the portfolio return afterward. For a young worker without leverage, ctc_tct​ is tiny in the early years because investable financial wealth is tiny in the early years. That means most of the effective risky exposure gets loaded into later contributions, when salary and savings are larger. So even if you think you are being conservative by avoiding leverage, you are implicitly making a large bet that the market environment in your later high-savings years will be favorable.

That is what I mean by greater tail risk. Your lifetime result is more exposed to a bad sequence occurring during the limited period when you finally have substantial money in the market. A moderate amount of leverage early can flatten this problem by shifting some exposure forward in time:

WT(L)=∑t=1TLtct∏u=t+1T(1+ru−fu)W_T^{(L)} = \sum_{t=1}^{T} L_t c_t \prod_{u=t+1}^{T} (1+r_u - f_u)WT(L)​=t=1∑T​Lt​ct​u=t+1∏T​(1+ru​−fu​)

where LtL_tLt​ is leverage and fuf_ufu​ is financing cost. If Lt>1L_t > 1Lt​>1 when young and then declines toward 1 or below later, you are effectively smoothing market participation across your working life. I.e you are amortizing your risk over time and this is actually a much SAFER position for you to be in financially.

Another way to formalize it is through variance of terminal log wealth. In a stylized model, if returns are independent over time, the variance of lifetime outcome depends not only on return variance σ2\sigma^2σ2, but on how contribution weights line up with those returns:

Var(WT)∝∑t=1Twt2σ2\mathrm{Var}(W_T) \propto \sum_{t=1}^{T} w_t^2 \sigma^2Var(WT​)∝t=1∑T​wt2​σ2

where wtw_twt​ is the effective contribution weight from each period. Without leverage, wtw_twt​ tends to be heavily back-loaded, because young investors have little capital. With prudent early leverage, the weights can become more evenly distributed. And because concentration raises variance through the sum of squared weights, a more even spread of lifetime exposure can reduce downside dispersion. The same idea appears in more advanced lifecycle models: the issue is not just “how much stock,” but when the stock exposure is taken.

This is also why the common objection “but leverage increases volatility” is incomplete. Of course it does at the portfolio level over short horizons. If you zoom in on one year, or one bad month, leverage mechanically magnifies mark-to-market moves:

rt(L)=Lrt−(L−1)rf−costsr_t^{(L)} = L r_t - (L-1)r_f - \text{costs}rt(L)​=Lrt​−(L−1)rf​−costs

So yes, one-year portfolio variance rises roughly like:

Var(rt(L))≈L2σ2\mathrm{Var}(r_t^{(L)}) \approx L^2 \sigma^2Var(rt(L)​)≈L2σ2

That part is trivial. But the relevant question is not whether leverage raises instantaneous volatility. It obviously does. The relevant question is whether a modest amount of leverage in the early years can reduce the probability of a bad retirement-level outcome by reducing the concentration of equity risk into a smaller late-life window. That is a very different question, and it is exactly where lifecycle leverage gets its force.

You can think about it in Samuelson-share terms. Suppose your target lifetime equity allocation is s. Then the young investor’s economically correct equity share of financial assets may be greater than 100% whenever human capital is large relative to financial capital:

Et∗Ft=s(1+HtFt)\frac{E_t^*}{F_t} = s\left(1 + \frac{H_t}{F_t}\right)Ft​Et∗​​=s(1+Ft​Ht​​)

If HtFt\frac{H_t}{F_t}Ft​Ht​​ is large enough, then Et∗Ft>1\frac{E_t^*}{F_t} > 1Ft​Et∗​​>1. That is not a mathematical oddity. It is the natural implication of thinking in terms of the whole balance sheet. A prohibition on leverage then acts like a hard constraint that forces the young investor below the exposure that would equalize risk over the life cycle. Recent lifecycle solution papers still analyze these leverage constraints explicitly because they materially change the optimum.

There is also a deeper intuition here that people miss. Not using leverage when young feels safe because account balances are low. But safety is partly an illusion created by small nominal stakes. The real risk is that you are postponing the bulk of your participation in the equity risk premium until later, when you have more capital and fewer years to recover. In that sense, the unlevered young investor is not necessarily less risky. It is just risk-delaying.

If the equity premium exists and if the investor’s labor income is not too equity-like, delaying risk can be costly. Ayres and Nalebuff’s work argues that moderate early leverage can increase expected retirement wealth while also improving downside retirement outcomes because it lets investors buy into the equity premium earlier and more evenly. Their point is not “max leverage all-in.” It is that a moderate, declining leverage schedule can create a more balanced lifetime exposure than either target-date conservatism or simply being 100% unlevered in stocks from day one.

Now, to be fair, this argument is not universally correct. Its strongest form depends on assumptions. The biggest one is the character of human capital. If your labor income is stable and mostly bond-like, then the case for taking more financial risk when young is stronger. But if your human capital is already highly correlated with the stock market—say you work in finance, venture, or a highly cyclical tech sector—then your total balance sheet may already be implicitly long equities. In that case, leverage can make your true exposure dangerously concentrated rather than better diversified. This is not a minor footnote; it is one of the major critiques in the literature on this subject. But for most of you wendy's dumpster operators, you can approve a bond like labor value asse