Forex as a Polish company (CIT) — when an LLC actually pays off

The most common question on Polish tax forums for traders runs roughly like this: "I'm booking 200,000 PLN a year on forex, my accountant friend mentioned a limited liability company would put me on 9% CIT instead of the 19% PIT-38 — is it worth it?" The honest answer requires a calm look at the full arithmetic, not just the nominal headline rates. In this article I show when a Polish sp. z o.o. (limited liability company) actually beats individual taxation, and when it is simply a more expensive version of the same outcome.

PIT-38 versus CIT — what we are really comparing

A Polish trader holding an account with a regulated broker (under KNF or another EU regulator — see the broader taxes and record-keeping reference) has three realistic paths for booking forex profits in 2026. The first is PIT-38, the personal capital gains tax — a flat 19% on net income (revenue minus deductible costs), no scale brackets, no mandatory social security contributions on this income, no full bookkeeping, no formalities beyond an annual PIT-38 return based on the PIT-8C report from the broker. The second is a sole proprietorship (JDG) on the lump-sum tax regime, which for most single-operator traders works out to around 15 to 17 percent of revenue plus ZUS social security. The third is a limited liability company (sp. z o.o.) under one of the corporate tax regimes — and this is the path that breeds the most illusions.

The headline CIT rate looks promising. In 2026, Poland runs three parallel CIT regimes: the standard 19% rate, the preferential 9% rate for small taxpayers (revenue up to the zloty equivalent of 2 million euro per year), and the so-called Estonian CIT regime (formally: lump-sum tax on company income), under which corporate tax only crystallises at the moment profit is distributed to a shareholder. Each regime has its own logic and its own real break-even point — and that is what decides whether a company is a sensible structure for a given trader or an expensive ornament.

Why 9% CIT is not actually 9% effective

The single most common error in this debate is comparing raw rates: PIT-38 at 19% versus CIT at 9%, with the breezy conclusion that "the company saves you ten percentage points". This is simply not true, because a company taxes the same profit twice. First the company pays CIT on its income, and then — when the profit is distributed as a dividend to the owner — the shareholder pays 19% withholding tax on the dividend at source. These two taxes combine into a single effective cost on the same zloty.

The arithmetic looks like this: a small-taxpayer company earns 100,000 PLN, pays 9% CIT (so 9,000 PLN), leaving 91,000 PLN. It then distributes all 91,000 PLN as a dividend, on which 19% is withheld at source (17,290 PLN). The owner receives 73,710 PLN in pocket. The combined tax burden is 26,290 PLN, an effective rate of roughly 26.3 percent — more than seven percentage points above the 19% PIT-38 path. Under the standard 19% CIT rate the picture is even worse: 19,000 PLN of CIT plus 15,390 PLN of dividend tax gives 34,390 PLN, an effective 34.4 percent on the same hundred thousand.

The conclusion is straightforward. If you plan to draw profits quarterly or yearly for ordinary living expenses, an LLC under classical CIT is simply a more expensive version of PIT-38. The company logic only starts to work when profits are retained for capital accumulation — in that case you redeploy 91,000 PLN instead of 81,000 PLN, and across a horizon of ten or more years the gap compounds non-linearly together with whatever return you earn on the retained capital.

Estonian CIT — when retaining profits actually pays off

The Polish Estonian CIT regime (formally: lump-sum tax on company income) was introduced in 2021 and substantially liberalised in 2022. The mechanics are clean. The company pays no corporate income tax on profits retained inside the firm. Tax only crystallises at the moment of distribution to a shareholder — at that point the rate is 10% for a small taxpayer or 20% for a larger entity, with the 19% dividend withholding tax on top and a partial credit for the corporate-level tax against the shareholder's personal tax liability. The effective combined burden on distribution lands somewhere around 20 to 25 percent, still slightly above the 19% PIT-38 figure, but with the crucial difference that retained profits are not taxed at all.

For an active trader compounding capital this is a structural shift. Imagine a trader booking an average of 100,000 PLN of profit a year over ten years and reinvesting every zloty back into the trading account. Under PIT-38 the trader pays 19,000 PLN of tax each year and has 81,000 PLN left to redeploy. Under Estonian CIT with full retention the trader pays nothing and redeploys the full 100,000 PLN. At a realistic annual return of 10 to 15 percent on retained capital, after a decade the first path reaches roughly 1.3 million PLN of accumulated equity, the second roughly 1.75 million PLN — a gap of around 400,000 PLN created purely by the deferral of tax.

"The Estonian CIT regime is the most significant change in Polish corporate taxation in twenty years. For active traders and individual investors planning to accumulate capital, it functions on a similar level to IKE or IKZE personal retirement accounts, with the important difference that there are no annual contribution limits." — Tomasz Krywan, tax advisor, Polish National Chamber of Tax Advisors, interview for Rzeczpospolita, March 2024.

Eligibility for Estonian CIT is precisely defined in the statute and requires careful analysis. The company must be a limited liability, joint-stock, or limited partnership entity. It must employ at least three people (with a meaningful relaxation for small companies in the first two years — a single full-time employee is sufficient), it cannot generate certain categories of passive income above set thresholds, and a substantial part of its assets cannot consist of financial assets. This last condition tends to be the awkward one for trading-focused entities, so consultation with a tax advisor experienced in the financial sector is mandatory rather than optional here.

VAT, ZUS and fixed costs — where the company eats the saving

The second layer of the calculation, the one nominal rates never show, is the annual cost of running the company itself. This is where the romantic vision of a "cheap company" built on headline tax rates usually falls apart. VAT happens to be the friendly part of the picture — trading CFDs on forex is exempt from VAT as a financial service (under Article 43 of the Polish VAT Act), so a trading-only company does not become an active VAT payer at all. The rest of the cost stack is less friendly.

The ZUS contribution of a sole shareholder is the most frequently overlooked line item in this whole conversation. Since 1 January 2022, the sole shareholder of a Polish limited liability company has been treated, for social insurance purposes, as a self-employed person and is subject to mandatory ZUS contributions. The standard workaround — introducing a second shareholder with a realistic stake of at least 10 to 15 percent — is now a routine advisory recommendation, but ZUS has increasingly challenged structures where the second shareholder is purely nominal. A symbolic 5% holding with no involvement in management and no dividend history can be reclassified as an artificial construction, and the financial consequences then reach back five years with interest.

The real break-even — where the company actually starts to win

The break-even point for a Polish sp. z o.o. flows from a simple piece of arithmetic: the tax saving must cover the fixed costs and still leave a visible margin. Let us run the numbers on three profit levels.

Below that threshold, the fixed costs of running the company eat the tax saving, and the additional weight of formalities (KRS filings, financial statements, monthly tax returns, JPK reports, ZUS deadlines) introduces stress with no economic upside. Above that threshold, and only when the strategy genuinely involves multi-year capital accumulation, the Estonian CIT regime becomes a real optimisation tool. The classical 9% CIT path with regular dividend distributions essentially never wins in practice.

Three paths — when each one applies

The most honest summary of a Polish trader's situation in 2026 looks like this. Up to roughly 150,000 PLN of annual forex profit the right form is the individual PIT-38 capital gains tax. Simple, cheap, no ZUS on this income, no full bookkeeping, no risk of structural disputes with the authorities. Between 150,000 and 250,000 PLN of annual profit a sole proprietorship on the lump-sum tax regime becomes worth comparing — cheaper than a company, formally simpler, and the 15 to 17 percent lump-sum rate on revenue often lands at a similar effective level to PIT-38. A limited liability company only enters the conversation above this band, and realistically only in the Estonian CIT variant combined with a genuine reinvestment strategy.

It is also worth remembering that losses from forex are accounted for differently under PIT-38 than inside a company — under PIT-38 a loss carries forward for five years and reduces taxable income from the same source, while inside a company the loss flows into the financial result and reduces the CIT base in subsequent years. The mechanics differ but the financial effect is similar, except that inside a company the fixed costs are incurred even in loss-making years.

Pitfalls that the spreadsheet does not show

The decision to incorporate has several consequences that a tax calculator never displays, and that frequently shift the verdict on the whole exercise. The first is public disclosure. The Polish court registry (KRS) is a public database — your details as a board member, the company's registered address, the share capital structure, and the annual financial statements (including the balance sheet and the profit and loss account) are openly accessible online. A portion of traders consider this a serious privacy drawback, particularly once annual profit becomes a publicly searchable figure.

The second is the reporting load: monthly or quarterly CIT advance payments, the annual CIT-8 return, financial statements signed by the board, JPK_KR accounting file submissions. Missing any of these deadlines triggers interest charges, and in serious cases administrative proceedings with the tax authority (KAS). Professional bookkeeping is not optional under this regime, it is a structural requirement. The third is the cost of exit — liquidating a Polish LLC takes between nine and eighteen months and runs to 5,000 to 15,000 PLN of fees for the liquidator, advisor and notary, plus a liquidation tax on assets transferred back to the shareholders. If you set up the company too early and find within a year that the structure was not worth it, the closing process can cost more than one year's tax saving.

The fourth pitfall is the risk of a dispute with ZUS or the tax authority over a nominal second shareholder. Bringing in a second shareholder purely to avoid the mandatory ZUS contribution of a sole shareholder has become common enough that the authorities now actively challenge such structures as artificial. If the second shareholder holds 5 percent, has no involvement in management, and never receives a dividend, ZUS may reclassify the structure as a sham, reach back five years for contributions, and add interest — the total exposure can comfortably exceed 100,000 PLN. The recommended minimum is a second shareholder with at least 10 to 15 percent of the equity and a genuine operational role.

What to do tomorrow before you call the notary

1. Write out, by hand, the last three years of your forex profits, year by year, in zloty, without optimism and without the "maybe this year will be better" reflex. If you have cleared 200,000 PLN net in at least two of those three years and you can see a realistic path to sustaining that scale for the next two or three years, a company may be worth considering. If the results are volatile, wait another year — a company formed after a single good year is the riskiest decision in this entire conversation.
2. Open a spreadsheet and model three scenarios on your real numbers: PIT-38 as the baseline, JDG on the 15% lump-sum tax, and an sp. z o.o. under Estonian CIT with full reinvestment. Include 35,000 to 45,000 PLN of fixed annual costs for the company, the ZUS and health-insurance contributions, and the potential audit. Only these three concrete numbers, not the headline rates, show which path actually wins in your situation.
3. Budget 500 to 1,500 PLN for a consultation with a licensed tax advisor (KIDP) experienced in the financial sector — you can find the directory on the chamber's website. Present your modelled scenarios and ask the advisor to validate the assumptions (activity classification, treatment of deductible costs, Estonian CIT eligibility for a trading-focused entity). The cost of that consultation is negligible against the consequences of an incorrect decision that runs for the next five years.
4. If the decision does come out in favour of incorporating, find an accounting firm with a documented track record of running trading and investment companies. Not every bookkeeper understands the specifics of forex reporting from foreign brokers, currency conversions, CFD contracts, commissions and overnight swaps. The right firm will cost 800 to 1,500 PLN per month, and it is one of the best-spent line items in the entire structure.
5. If your annual profit is below 150,000 PLN, set the company question aside for now and focus on diligent use of deductible costs under PIT-38 — broker commissions, analytical service subscriptions, VPS hosting, properly invoiced education, computer equipment used for trading. Every 10,000 PLN of deductible costs produces 1,900 PLN of real tax savings. No formalities, no fixed costs, fully legal, and entirely sufficient for this scale of activity.