What Actually Makes an Estate Plan Better

Most estate plans look about the same on paper. A revocable trust, a pour-over will, a financial power of attorney, a health care directive. The signing appointment feels the same no matter who drafted the documents. The differences do not show up for years, and when they finally do, the person who could have fixed the problem has died.

So when people ask what separates a good estate plan from a mediocre one, the answer is not the paper. It is whether the plan was written by someone who has watched plans come apart.

Here are three things I look for, and one I think matters more than all of them.

The worst feature a married couple's trust can have is that it locks shut when the first spouse dies

For decades the standard plan for married couples was the A/B trust, also called a bypass, credit shelter, or exemption trust. On the first death, the trust splits. The survivor's half goes into the Survivor's Trust, which stays revocable. The deceased spouse's half drops into the B trust, which becomes irrevocable immediately and permanently.

That structure existed for one reason: to preserve the deceased spouse's federal estate tax exemption. Before 2011, if you did not use that exemption at the first death, you lost it. The mandatory split was the workaround.

That reason is gone. Portability now lets a surviving spouse carry over the deceased spouse's unused exemption, and the federal exemption sits at $15 million per person in 2026, made permanent and indexed for inflation by the 2025 tax act. California has had no estate tax since 1982. For the overwhelming majority of married couples in this state, a mandatory bypass trust now buys nothing at all in tax savings while charging full price in every other way.

What the survivor gets instead is a second trust they cannot change, with its own tax identification number, its own Form 1041 every year, its own accounting obligations to the children as remainder beneficiaries, and a trustee's duty to treat those children as beneficiaries with enforceable rights in property the survivor thought of as theirs. Trusts also hit the top federal income tax bracket at a level of retained income that would barely register on an individual return.

Then there is basis. California community property gets a full step-up in basis on both halves at the first death. That is a genuine advantage of living here. But the assets sitting in the B trust get that adjustment once and never again. When the survivor dies twenty years later, those assets carry their old basis while everything in the Survivor's Trust steps up a second time. On a house in California, the difference in capital gains tax to the children can be enormous, and it is being paid to solve an estate tax problem the family never had.

One more thing about portability that gets missed: it is not automatic. The estate has to file a federal estate tax return to elect it, even though no tax is owed. Plenty of surviving spouses lose the exemption because nobody told them a return was needed.

Fixing a locked trust after the fact is expensive, and the survivor pays for it

I see this regularly. A surviving spouse comes in with a trust drafted in 1998, discovers that half of everything now belongs to an irrevocable trust, and wants to know how to undo it.

Sometimes you can. Probate Code section 15409 allows a court to modify or terminate an irrevocable trust when circumstances the settlors did not anticipate would defeat or substantially impair the trust's purpose, and the collapse of the estate tax rationale is a reasonable fit. Section 15403 allows modification with beneficiary consent. If the instrument gives the trustee enough discretion, there may be a way to distribute the bypass assets out without a court at all.

But consent has to come from all the beneficiaries, including contingent ones, which in practice means the children and sometimes grandchildren who are not yet born. Minor and unborn beneficiaries cannot consent, so you are looking at a guardian ad litem. That means a verified petition, notice to everyone, a hearing, and often several months. The legal fees are real money, and they come out of the same estate the plan was supposed to protect. If one child does not sign off, you have a contested proceeding, and the family relationship goes with it.

All of that is the cost of a drafting decision made decades earlier that nobody revisited. The survivor inherits the bill.

The disclaimer trust gets you the same options without forcing anyone's hand

The better structure for most couples is a disclaimer trust.

Everything passes to the Survivor's Trust by default. The survivor keeps full control, one trust, one tax return, no split, no accountings to the kids. But the trust also contains a fully drafted bypass trust that sits dormant. If the survivor decides, after the first death, that some or all of the deceased spouse's share should go into an irrevocable trust, they disclaim it within nine months and it funds automatically.

The advantage is timing. The decision gets made when the numbers, the tax law, the survivor's health, the creditor picture, and the family situation are all actually known, instead of being guessed at years in advance by people who could not have predicted any of it. If none of those concerns materialize, the survivor does nothing and the plan stays simple.

Two things have to be right for this to work, and they are where sloppy drafting shows.

First, a disclaimer only counts if it is a qualified disclaimer under Internal Revenue Code section 2518 and Probate Code sections 275 through 285. It has to be in writing, within nine months, before the survivor accepts any benefit from the property, and without the survivor directing where the property goes. That last requirement is why the disclaimer trust's terms have to be written out completely and fixed in the document itself. The survivor cannot pick the terms afterward. A document that lets the survivor decide what the trust says has destroyed the disclaimer.

Second, someone has to tell the surviving spouse the nine month window exists. A disclaimer plan requires an affirmative act during the worst year of a person's life. That is an argument for a trust administration meeting after the first death, not an argument against the structure.

There is one situation where the mandatory version still makes sense. In a blended family, where the whole point is to guarantee that the first spouse's share reaches that spouse's children and cannot be redirected by the survivor, locking the share at the first death is the feature, not the bug. The difference is that it should be a deliberate choice made for that reason, not a leftover from a tax rule that expired.

If there is significant separate property, the answer is a separate trust, not an A/B split

This is where I see the most damage.

An A/B trust divides property by tax formula. It does not divide property by character. Feed a couple's community property and one spouse's inherited separate property into the same joint trust, run it through a bypass formula at the first death, and you have buried the separate property question instead of answering it. Years later, when the survivor has died and the children from the first marriage are asking what happened to their mother's inheritance, nobody can trace it.

When one spouse brings substantial separate property to the marriage, or inherits during it, the cleanest structure is a separate trust for that property alongside the joint trust. The separate property stays identifiable. Tracing stays possible. You avoid the argument that transferring separate property into a joint trust was a transmutation into community property, which under Family Code section 852 requires an express written declaration but which litigants argue about anyway. And when the property eventually passes, it passes under terms written specifically for it, rather than under a formula designed for a tax that no longer applies.

Two trusts sounds like more complexity. It is usually less, because it answers the hard question on the front end rather than leaving it for a court.

The thing that matters most is whether your attorney has seen plans fail

Estate planning is one of the few areas of law where the work product is not tested until the drafter is long out of the picture. You can write documents your entire career and never find out which of your choices caused problems.

I have been on the other side of that. I have handled trust and will contests, petitions under Probate Code section 17200, claims over property ownership, disputes over capacity and undue influence, and fights between a surviving spouse and the children from a prior marriage. That experience changes how you draft, in ways that are hard to get any other way.

It is the difference between including a no contest clause and knowing how narrowly California courts enforce them, and what actually has to be in the document for one to matter. It is knowing which trustee provisions turn into deadlock and which successor language leaves a gap. It is knowing that the phrase "as the trustee deems appropriate" is where litigation lives. It is knowing when a client's plan is going to draw a challenge, and building the record now, while the client is here to make it, rather than hoping the file holds up later.

A plan that is cheap to sign and expensive to administer is not a good deal. The measure of an estate plan is not what it costs today. It is what it costs your family when it finally has to work.

Steven J. Chamberlin is a California attorney whose practice includes estate planning and trust and probate litigation. If you have an existing plan with a mandatory A/B split, or separate property mixed into a joint trust, it is worth a review.

This article is general information about California law, not legal advice, and no attorney-client relationship is created by reading it.